1 FOREWORD

Welcome to the 10th AVL International and transmissions to reduce CO2, and what is the Commercial Powertrain Conference – ICPC, a potential of vehicle and machine electrification to platform to discuss common challenges and achieve zero emissions. opportunities of truck, agricultural and construction equipment OEMs, organized by AVL in Digitalization is the game changer, both on-road cooperation with SAE International. and off-road.

The on-road and off-road industry is in rapid The closing round table discussion will try to find transition: to meet ever stringent environmental, answers if the Commercial Powertrain Industry will legislative and economic targets, the current continue with steady evolution or be threatened by powertrain technology will no longer suffice. disruption with severe implications on the companies’ success, job security and the The Keynote session, with the roadmap to drive the economy. future and the implications on logistics, environment, and economics sets the scene for the We are once again looking forward to an exciting conference. event, an opportunity to meet and network with many representatives of the leading companies of Experts from the three powertrain industries will tell the three powertrain industries including their us what lies ahead in powertrain development, strong supplier base. what can still be achieved with combustion engines

Dr. Marko Dekena Prof. Dr. h.c. Helmut List DI Rolf Dreisbach Executive Vice President Chairman and CEO Executive Vice President Off-road Business Unit On-road Business Unit

ICPC 2019 May 22nd - 23rd, 2019 2 CONTENT Page

ON-ROAD AND OFF-ROAD INDUSTRY IN TURMOIL?

1.1 Roadmap to drive the future - 8 The powertrain diversification

Matthias Horx – Founder; Zukunftsinstitut Horx GmbH

1.2 Technologies and Infrastructure needed for Sustainable Logistic 11

Prof. Dr. Helmut Zsifkovits - Professor of Industrial Logistics; University of Leoben

1.3 What is needed to ensure efficient and clean road transport? 19

Dorothee Saar - Head of Transport and Air Quality; Deutsche Umwelthilfe e.V

1.4 Economic Implications: 22 Impact of e-mobility on supply chain and related market expectation

Gerhard Stempfer - Manager Elektrification; ZF Friedrichshafen AG

3 CONTENT Page

INNOVATIVE ICE AND EAS SOLUTIONS

2.1 Innovative EAS technologies in development 26 for on-road and off-road applications

Rolf Brück - Managing Director; Continental Emitec GmbH

2.2 Lowest CO2 Emissions Despite Ultra-Low NOx 28

Gernot Graf - Head of Development & Calibration, Commercial Vehicle and Large Engine, Helmut Theißl, Klaus Hadl, Anton Arnberger; AVL List GmbH

2.3 Development and testing of an innovative gas engine for heavy 36 duty applications Stefano Golini - Alternative Fuels Project Manager – Innovation Dept., David D’AMATO, Sergio GIORDANA, Paolo GROSSO, Diego IUDICE; FPT Industrial S.p.A. Anton ARNBERGER, Gernot HASENBICHLER; AVL List GmbH Davide PAREDI; Politecnico di Milano Peter GRABNER; Technische Universität Graz

2.4 Potentials for friction reduction with commercial vehicle engines – 56 Contribution of the power cell unit

Dr.-Ing. Andreas Pfeifer – Vice President Product Development Engine Systems; Mahle GmbH M.Sc. Tobias Funk, Dr.-Ing. Thomas Deuß, Dipl.-Ing. Holger Ehnis MAHLE International GmbH

2.5 ICE optimized for off-road hybrid powertrain 63

Dr. - Ing. Markus Schwaderlapp - Senior Vice President Research & Development, Dr.-Ing. Paul Grzeschik; DEUTZ AG

4 CONTENT Page

VEHICLE AND MACHINE ELECTRIFICATION

3.1 Fuel Cells: 71 A Profitable Zero-Emission Solution for Heavy Duty Trucks

William Resende, Global Product Manager Fuel Cells, Heimo Schreier, Dr. Alexander Schenk, Martin Ackerl; AVL List GmbH

3.2 Commercial vehicle battery solutions 82

Krzysztof Paciura - Technical Project Lead (Power Electronics R&D/Project Manager); Cummins Ltd.

3.3 The eCVT hybrid system for 2020+ Commercial vehicle 87 application

Dr.Zhiqiang Lin - Executive Vice President; Guangxi Yuchai Machinery Co., LTD

3.4 Tractor/implement systems – the next generation 92

Dr.-Ing. Joachim Sobotzik - Versatility Tractor Lead, Electric Drive Engineering Services; John Deere European Technology Innovation Center Kaiserslautern, Germany

5 CONTENT Page

CONTRIBUTIONS TO REDUCE CO2 - ON-ROAD AND OFF-ROAD

4.1 The initiative of the AG machinery industry for CO2 emission 96 reduction

Dr. Eberhard Nacke - Head of Product Strategy; Claas KGaAmbH

4.2 Long haul truck powertrain control for low emission and fuel 99 consumption in real traffic conditions

Alois Danninger – Manager Engine Controls; AVL List GmbH

DIGITALIZATION - THE GAME CHANGER, ON-ROAD AND OFF-ROAD

5.1 Impacts of Digitalization on the Ag Industry 106

Dr. Markus Baldinger - Chief Technical Officer; Pöttinger Landtechnik GmbH

5.2 ADAS and Autonomy for Trucks - A look into the future 109

Ozan Nalcıoğlu - Engineering Director, Alper Tekeli, Güvenç Barutçu, Berzah Ozan; Ford Otomotiv AS

5.3 Digital Twins - Prerequisites and implications on 115 CE architectures and platforms

DI Dr. Thomas Fischinger – Digitalization Consultant Wacker Neuson Beteiligungs Gmbh

6 CONTENT Page

COMMERCIAL POWERTRAIN INDUSTRY: EVOLUTION OR DISRUPTION?

6.1 Powertrain Trends and Developments in the Commercial Vehicle 119 Industry

Dr. Carl Hergart - Director Powertrain and Advanced Engineering; PACCAR Technical Center

6.2 Value Creation in the Commercial Powertrain Industry 128

Dr. Albert Neumann - Managing Director, Jana Mühlig, Xaver Müller; Strategy Engineers

7

ICPC 2019 – 1.1 Roadmap to drive the future - The powertrain diversification

Matthias Horx Zukunftsinstitut Horx GmbH

Copyright © 2019 AVL List GmbH, Zukunftsinstitut Horx GmbH and SAE International

ABSTRACT wenn sie alkoholmässig über die Stränge schlagen (das kommt schon mal vor, wird allerdings immer When will the transportation business reach the next weniger, seit die isländische Regierung strenge transformation stage, and what will be the factors and Gesundheitsprogramme ins Leben gerufen hat). trends which lead to this point? There are three forces Wenn sie genug gesammelt haben, geben die Trolle which shape the coming shift: die Bierdosen beim nächsten Recyclinghof, auf isländisch endurvinssla stofnun ab. Deshalb sieht Energy and environmental: Island so unglaublich sauber aus, wie geleckt.

The Anti-Global-Warming -Movement will become a Trolle sind normalerweise sehr verspielt. Wenn man powerful force in the next decade, and shape value mit einem Geländewagen - wir waren zum Beispiel systems and cultural behaviour. The goods mit einem alten Landrover Defender, mindestens transportation system will come under huge pressure. zwanzig Jahre alt, das Modell, das noch ganz ohne Electrification or hydrogen tech will come sooner then Federung auskam unterwegs - auf einer der endlosen we think. Geröllstrassen durch die grandiose Landschaft fährt, schmeissen sie unablässig kleine Steine an die Technology and Infrastructure: The steadily increasing transportation traffic creates Windschutzscheibe. Und kichern. Das hört man aber tensions between different groups on the road. But kaum, wegen des Fahrlärms. automatic driving systems cannot change congestion, so we need more radical proposals such as: Trolle können allerdings wachsen. Ziemlich schnell. Und je grösser sie werden, desto schlechter wird ihre "The Human Factor": Laune. Sie machen dann alles Mögliche kaputt. Zunächst noch aus Neugier, einfach mal so, um zu Lorry driving is a very mighty and deeply rooted male schauen, was passiert. culture. At the same time lorry drivers are suffering under hard economic pressures. With the coming Irgendwann sind sie so gross wie eine Kuh. Oder ein technological shifts, this will lead to conflicts around riesiger Felsbrocken. Oder ein kleiner Berg. Und sie the steering wheel. Another driver culture with higher verhalten sich dann - wie sagt man so schön? - educational skills will emerge. ungünstig. Wie ein Erdbeben, das man lieber nicht erleben möchte. DAS TROLL-PRINZIP Weglaufen nutzt dann nichts mehr. Vor einiger Zeit reiste ich mit meiner Familie nach Island, in dieses aussergewöhnliche Land der Natürlich glauben wir nicht an solche Geister- Geysire, Vulkane und fantastischen Nordlichter. Dort Geschichten. Das sind Märchen für Kinder, wie etwa wohnen, wie jeder weiss, die Trolle. die übelgelaunte Gr´yla, eine alte Frau die in der dunklen Zeit vor Weihnachten aus den kalten Bergen Trolle sind eigentlich harmlose, sogar nützliche kommt und bis zu 10.000 auf einmal Kinder frisst. Wesen. Sie hausen in Spalten im losen Lava-Geröll; Jedes isländische Kind kennt sie. Oder die dort, wo Moos und Steine kleine Höhlen und Gänge berühmten Elfen, wegen denen in Island auch schon bilden. An den Strassenrändern sammeln sie leere mal eine Strasse umgeleitet wird. Island wimmelt von Bierdosen, die die Isländer, gelegentlich wegwerfen, solchen Sagen. Wie kann es auch anders sein in

8 ICPC 2019 – 1.1 einer Landschaft, die bis zum Weltraum reicht und Hässliche Dinge, die man nicht erzählen möchte. voller Wunder ist? Über weite Strecken besteht Island Aber was nutzt das, wenn man es weiss? aus marsianischer Öde ( Im nördlichen Teil hat ESA Mars, die Mars-Expeditons-Vorbereitungs-Mission Oder man kennt seine Nachbarin seit vielen Jahren. der europäischen Weltraumorganisation, ihre Eine lebensatte Frau, eigentlich. Bis vor einigen Kuppeln und Container aufgeschlagen). Wer in einer Jahren immer fröhlich und positiv und ziemlich solchen Einöde lebt, MUSS sich mit Elfen und Trollen katholisch. Plötzlich verändert sie sie sich, wird und bösartigen Hexen umgeben. langsamer, stummer, in sich gekehrter. Und dann kommt sie über den Gartenzaun mit allen möglichen Das Erstaunliche ist nur, dass Trolle, obwohl sie nur konfusen Theorien - über geheime korrupter Politiker- Imagination sind, äusserst REAL sein können. Kreise und die Manipulation des Wetters, den Islam als grosse Verderben. Auf ihrem Gesicht ist plötzlich Trolle sind Saboteure auf unserem Weg in die ein seltsame Genugtuung eingekehrt (Sie hat gerade Zukunft. Sie stehen sozusagen zwischen uns und eine ziemlich schlimme Scheidung von ihrem Mann, dem Kommenden, das möglich ist. einem Rechtsanwalt, hinter sich).

Sie leben von einer menschlichen Grundenergie: Die Angst scheint eine geheimnisvolle innerliche Unserer Angst. Angst ist ihre Lieblingsspeise. Sie Stärke zu verleihen. verzehren Angst zum Frühstück, zum Mittagessen und zu Abendessen, und gerne noch als kleine Am irritierendsten sind diese inneren Troll-Anfälle, Happen zwischendurch. Angst ist das, was sie gross, wenn sie sich bei Menschen bemerkbar machen, die stark und böse macht. Die Aufmerksamkeit, die die Zukunft geradezu für sich beanspruchen. Ich Fixierung, die durch Angst entsteht, ist das, was sie haben ein ums andere Mal erlebt, wie CEOS, Inhaber zu gigantischer Form auflaufen lässt, bis sie den grosser Firmen, charismatische Männer, ihre innere Himmel verdüstern. Dunkelheit offenbarten. Bei einer festlichen Kundenveranstaltung, nach einer flammenden Wir alle kennen diese Situation. Wir sitzen zusammen Motivationsrede zur gloriosen Zukunft ihrer Technik, mit einer Gruppe von Freunden, die wir kennen und ihres Unternehmens, sass ich beim Essen mit ihnen schätzen. Etwa in einem Gasthaus in den Bergen; zusammen. Und hörte beim obligatorischen jemand hat Geburtstag, man hat sich länger nicht Filetsteak, das bei solchen Anlässen immer gereicht gesehen, grosses Hallo zu Beginn. Man kommt ins wird, plötzlich diese apokalyptische Geschichten. Gespräch mit einem alten Bekannten, den man noch aus der Studentenzeit kennt, und mit dem man vor Sie als Zukunftsforscher geben mit doch sicher recht, Jahrzehnten einmal auf einer turbulenten Reise in dass... den Süden war, als man sich noch Hals über Kopf verlieben konnte. - Die Ressourcen zu Ende gehen (ausgerechnet die, mit dem die Firma des jeweiligen Konzerns ihr Geld Man sitzt und redet und trinkt ein bisschen Wein, nicht verdient)... mehr so viel wie früher. Und nach einigen schönen Erinnerungen formuliert der gute alte Freund plötzlich - Der Mensch sich wie wahnsinnige Kaninchen Sätze wie: vermehrt, das kann ja von den Ökosystemen her nicht gutgehen... „Ich glaube, es geht zu Ende. Das fliegt uns alles um die Ohren. Ich schätze, es geht noch zehn, zwanzig Der Euro demnächst auseinanderfliegt, diesmal aber Jahre... aber dann... Meine Kinder beneide ich ganz sicher... nicht.... “ Der Terrorismus unsere Gesellschaft zerstört. Schweigen. Verlegenes Lachen. Man versucht zaghaft einige Gegenargumente. Lieber X, haben wir Alle Liebe und Bindung, Anstand und Respekt das nicht in unserer Jugend auch schon so gedacht, zwischen den Menschen verschwindet. Du erinnerst Dich, Atomkrieg, Waldsterben, Kapitalismus, Tschernobyl? Aber der düstere Angst- Es gibt demnächst einen grossen Krieg, einen Glanz in den Augen des alten Freundes hat eine weltweiten Bürgerkrieg, bei dem die Armen die spezifische Hartnäckigkeit. Ja sogar ein bestimmtes Reichen abschlachten. Vielleicht ist das ja sogar Glück. gerecht...

Man kennt ihn ein bisschen. Man weiss: Es ist in Wir werden von Algorithmen versklavt und in einem Wahrheit eine Depression. Er hat, wie so viele der neuen Faschismus aufwachen, ohne dass wir es alten Rebellen, eine schwierige Kindheit gehabt. überhaupt merken....

9 ICPC 2019 – 1.1

Die letzte Variante wurde mir bei einem schweren Bei der Angst geht es immer auch um Macht. Oder Rothschild-Wein ausgerechnet vom Vorstand einer kompensierte Ohnmacht. grossen IT-Firma aufgetischt. Mit einer Mischung aus Zynismus und eitler Selbstgeisselung ("Wenn Sie Es ist leicht, sich von den elektronischen Hass- glauben, ich wüsste nicht, was wir tun, dann irren Sie Trollen zu distanzieren. Das Pathologische ist sich!"). offensichtlich. Aber es macht wenig Sinn, die Existenz der Trolle einfach zu negieren. Die Trolle Das Bizarre an diesen Ängsten ist ihre seltsame sind seit Millionen von Jahren mit uns gewandert. Beliebigkeit ; meistens bewegt sich die Narration Jetzt, im Zeitalter der elektronischen Verbindungen entlang von uralten Schauergeschichten, die schon und Erregungen, werden sie endgültig aus ihrem seit eineigen Jahren völlig widerlegt sind (etwa die Schlaf geweckt. Sie können, wie wir gesehen haben, "Bevölkerungsexplosion" oder die "ständige sogar das Amt des amerikanischen Präsidenten Zunahme der Arbeitslosigkeit"; beliebt ist auch die bekleiden. "Vergreisung der Gesellschaft", die einfach nicht innovativ und digital genug ist, um es mit der Zukunft Die Erde zittert, das können wir spüren. "aufzunehmen".). Geschichten, die irgendwo aus der Tiefe des medialen Bauches stammen, und unendlich Es wird also Zeit für einen Showdown mit dem zu weitergesponnen werden, wie im Märchen von gross gewordenen Troll. Rapunzel das Haar. Nicht selten offenbart sich in der Art und Weise, wie die Untergangsgeschichten erzählt werden, direkt eine persönliche Niederlage, ein frisch vernähtes Trauma. Aber immer laufen sie auf dasselbe hinaus:

ES GIBT KEINE ZUKUNFT!

Wenn alle, oder die Mehrheit, das wirklich glaubt, dann ist der Moment gekommen, in dem der grosse, der ganz grosse Troll den Raum betritt, sich genüsslich am Hintern kratzt, um sich dann sich ächzend, stinkend und dennoch sehr gemütlich niederzulassen.

Es ist dann ziemlich schwer, ihn wieder zu vertreiben.

Das Internet hat dem Troll inzwischen eine neue Ikonographie verliehen. Das grinsende Clownsgesicht, das unentwegt ablacht - die Maske des Bösartigkeit und hämischen Besserwisserei, mit der jeder irgendwann konfrontiert wird, der in die Öffentlichkeit tritt. Gerade in dieser Figur des Internet- Trolls zeigt sich das Paradoxon der Angst: Der Seelengenuss, der von reiner Negativität auszugehen scheint.

Die Psychologie der Internet-Trolle ist inzwischen weitgehend erforscht. Es handelt sich zum grossen Teil um Menschen mit einem fatalen Selbstwirksamkeits-Problem. Der Mann (90 Prozent sind Männer), der in einem abgeschlossenen Raum vor einem Computer sitzt, und aus der komfortablen Distanz des elektronischen Raumes andere Menschen belästigt, beleidigt, verhöhnt, verunsichert, denunziert, verfolgt, nimmt eine Art Superposition ein. Er kann die Gefühle anderer in Richtung Angst manipulieren. Das gibt dem Ohnmächtigen eine ungeheure Wirksamkeitserfahrung. Eine Kontroll- Illusion. Das ist der Hebel seiner Wirksamkeit; es stabilisiert ihn in seiner eigenen Lebensangst.

10

ICPC 2019 – 1.2 Technologies and Infrastructure needed for Sustainable Logistic

Prof. Dr. Helmut Zsifkovits Montanuniversitaet Leoben

Copyright © 2019 AVL List GmbH, University of Leoben and SAE International

ABSTRACT production systems. These are measures of supply chain efficiency, like operating cost, lead time, The way goods and physical objects are currently flexibility and visibility, customer-focused indicators moved, handled, stored, distributed and supplied is like service level and solution portfolio. not sustainable economically, environmentally, and socially. Several examples will be presented to Furthermore, sustainability has become a critical illustrate the symptoms and effects of unsustainable factor in every logistics-related activity. The 2030 logistics. Based on a Smart Logistics framework, Agenda for Sustainable Development of the United approaches towards more sustainable transport, Nations underlines a global commitment to “achieving storage and provision of goods are discussed. sustainable development in its three dimensions - Furthermore, a number of conceptual, technological economic, social and environmental - in a balanced and organizational models for Smart Logistics are and integrated manner” [2]. In order to achieve identified, and potential applications will be outlined sustainable development, i.e. “protect the planet from and discussed. degradation, including through sustainable consumption and production, sustainably managing INTRODUCTION its natural resources and taking urgent action on climate change, so that it can support the needs of the present and future generations”, the merging of Infrastructure comprises the entirety of sustainable the three dimensions in the context of decisions at facilities and supply channels to be used by private enterprise level and into the public policy cycle is households and companies. Economic/technical required. infrastructure consists of transport infrastructure, information and communication infrastructure Transport is a major user of energy and burns most whereas social infrastructure includes institutions for of the world's petroleum, therefore the environmental education, healthcare, culture and security within a impact of transport infrastructure and operations is state and society [1]. significant. Road transport is the largest contributor to global warming, creating air pollution, including A further distinction can be made between a nitrous oxides and particulates, and emission of macroeconomic view, as described above, and the carbon dioxide [3]. microeconomic dimension defining the technical structural properties of a logistics system, such as In 2015, the 2030 Agenda for Sustainable means of transport and material handling, conveyors, Development was adopted by all United Nations warehouses, storage and picking technology, and the Member States, as a “shared blueprint for peace and information and communication systems required for prosperity for people and the planet, now and into the controlling these facilities. future”. The agenda defines 17 Sustainable Development Goals (SDGs), calling for joint action by Logistics infrastructure forms the backbone of all countries to end poverty and other deprivations, to logistics systems, including the transport improve health and education, reduce inequality, and infrastructure, and the suprastructure including spur economic growth. Also, climate change has to logistics locations and real estate as well as the be tackled, and the preservation of the natural telecommunication infrastructure. environment [4]. Consequently, infrastructure has a major impact on the key performance indicators of logistics and

11 ICPC 2019 – 1.2

Resilient infrastructure is explicitly addressed in Goal It has to be taken into account, though, that 9 of SDG: “Build resilient infrastructure, promote infrastructure, as defined by highways, airports, inclusive and sustainable industrialization and foster power plants and dams does affect, and is affected innovation.” [5] by the environment. The construction and operation of infrastructure accunts for approximately 70 percent Resilient infrastructure is interlinked with of greenhouse gases (addressed in Goal 13). Linear industrialization and innovation which are drivers of infrastructure and dams have major impacts on infrastructure development and economic growth. terrestrial and aquatic ecosystems and biodiversity Thus, infrastructure is critical for achieving the (Goals 14 and 15). In contrast, the increasingly visible socio-economic SDGs, inclusive growth (Goal 8), effects of climate change, land degradation and reducing poverty (Goal 1), eliminating hunger (Goal deforestation pose major threats to infrastructure [5]. 2), ensuring good health and well-being (Goal 3), providing quality education (Goal 4), supplying clean The United Nations Economic Commission for water and sanitation (Goal 6). Europe (UNECE) was set up in 1947, with its major aim is to promote pan-European economic Access to energy, clean water and sanitation is integration, as one of five regional commissions of the central for better education and for better health, United Nations. UNECE goals with respect to and is also critical for gender equality as it increases infrastructure are to identify main Euro-Asian links for mobility, output and productivity of women, in priority development and cooperation, ensure particular. seamless connections throughout Europe through motorways and railways and facilitate border crossings.

Observatory – online repository for transport network information [6]

An online repository (Observatory, Figure 1) will be not sustainable economically, environmentally, and established to store and exchange basic information socially. The basic experience of transportation has on identified transport networks. All international not changed very much over the past 50 years. corridors will be hosted in a GIS environment [7]. Whereas other industries have undergone revolutionary changes, transport infrastructure and This will be a major foundadtion for providing a vehicles (cars, trains, planes) still move at roughly the comprehensive informational basis for future same speed as decades ago. The potential of infrastructure planning of transport networks. technological advancements has been compensated by the increasing density of frequency of moving The way goods and physical objects are currently objects – and regulation. Several examples will be moved, handled, stored, distributed and supplied is

12 ICPC 2019 – 1.2 presented to illustrate the symptoms and effects of optimize operations, in terms of efficiency, flexibility unsustainable logistics. and sustainability. Expected effects will be a more efficient use of shared resources, like storage space, • According to various surveys, between 20 and vehicles, and production facilities, and a multi- 40 percent of trucks run empty (without freight). echelon inventory optimization. This in turn enables In addition, there is a high percentage of partially more productive and economic operations, higher loaded vehicles. The inefficient use of trucks service levels, and more sustainable systems in their leads to too many trucks on European roads, economic, environmental and social dimensions. creates congestion and unnecessarily increases the direct and external costs (delays, damages, A SMART LOGISTICS FRAMEWORK injuries, fatalities, emission) of vehicles. In this section, the authors develop the framework for • The average passenger car is in use 1 out of 24 smart and sustainable logistics, outline and discuss hours. Some surveys even indicate, the vehicle state of the art conceptualizations and combine their is used only 36 minutes per day [8]. From this findings to a generic Smart Logistics Framework. fact, an immense requirement for parking space arises. Kagermann et al. (2013) define Industry 4.0 as the technological-based integration of Cyber-Physical • In Vienna, the area dedicated to cars is 12 million Systems (CPS) into production and logistics and the square kilometers; one fourth out of this is for application of the Internet of Things (IoT) and services parking cars. The ratio is similar in Berlin, in industrial processes, including all the resulting London and other major European cities. In consequences for the value chain, business models, Berlin, on average, only about 60,000 cars are services and labor organization. Production and traveling at one time, while 1.2 million are logistics are considered as core areas of Industry 4.0 parked. while business models and other corporate function will only play a minor role [9]. Smart Logistics is a core There is a number of approaches and initiatives, element of the Industry 4.0 concept. though, that indicate progress in sustainable supply chains. More logistics services providers are taking Hermann et al. (2016) derived four basic formal sustainability and resource consumption into account principles from relevant publications for the when planning or purchasing transportation. A turn implementation of Industry 4.0 approaches which are towards electric vehicles, and the usage of natural 1) the assurance of digital interconnectivity, 2) the gas for ground fleets are developments to be decentralization of decision-making processes, 3) the mentioned. Cleaner trucks, trains, ships and planes availability of transparent information, and 4) the are further factors, and the introduction of automation usage of technical assistance systems [10]. and cleaner technology in cargo handling and warehouse operations contribute to “greener” supply Bechtold et al. (2014) created a framework which chains. comprises eight important value drivers based on the four pillars of Industry 4.0, namely, 1) Smart Packaging is becoming more sustainable, due to the Solutions, 2) Smart Innovations, 3) Smart Supply usage of reusable and recycleable containers and Chains and 4) Smart Factory. In this model, products materials, and the reduction of wasted space. It must and services respectively innovation processes are not be neglected, though, that the use of boxes for E- conceptualized as independent elements. This Commerce is growing considerably faster than other framework should be used to ensure continuous market segments. growth and enhanced efficiency based on an organizational framework which includes an agile In order to eliminate or minimize wasteful resource operating model, human resource management, usage before it is even introduced, sustainability in change management, corporate governance, supply chains must be taken account of all the stages processes and digital infrastructure [11]. of planning and implementation, from product and process design to the sourcing of facilities, tools, Based on the outlined models, the authors propose a materials, components, energy and packaging to the new framework of Smart Logistics which is displayed distribution of goods. in Figure 2. Thereby, Smart Logistics includes intelligent and smart supply chains, based on an agile Digitization of supply chains can contribute to make cooperation in interlinked networks and the digital them leaner, help eliminate or minimize waste and interconnectivity of organizations. The digital avoid obsolencence. Data analytics, the use of interconnectivity is assured by state-of-the-art advanced algorithms and Artificial Intelligence information and communication technologies (ICT), provide tools to model and evaluate scenarios and data networks, sensors and actors, and intelligent

13 ICPC 2019 – 1.2 technologies for identification respectively tracking of infrastructure enable the complete or at least partial materials, components and/or products. Autonomous self-control of internal and external material flow transport vehicles in combination with automated processes [12]. warehouses as well as storage and handling

A smart and sustainable logistics framework (adapted from [12])

Smart Solutions are implemented through Smart products without human participation is possible, to Products and Smart Services. Smart Products based improve logistics processes, competitiveness and on Product-Enabled Information Devices (PEID), reinforce the confidence of consumers in the quality such as RFID, sensors, actors, allow the of the supply chain. Logistics systems efficiency can interoperability of systems by dynamically be increased by internally implementing real-time exchanging product data and additional in-depth planning and control systems and externally information related to lifecycle management, synchronizing information by using cloud-based products, and manufacturing processes. New product approaches or learning algorithms [14]. and process innovations may result from these. Fitting products with sensors can lead to improved Automated guides vehicles (AGV) become more and information for the development of new features or more important for industrial enterprises, with the materials, a better utilization and ongoing ability of reconfiguration, flexibility, and optimization of production and logistics systems customizabilityMoreover, warehousing processes are through real-time condition monitoring. supported by autonomous robots, intelligent carriers and advanced assistance systems for man-machine- Smart Innovations can be realized by developing interaction. cooperative innovation strategies and open innovation platforms which could be more effective in Smart Manufacturing uses data analytics to increase transforming company-internal innovation processes production efficiency by gaining deeper insights into into dynamic processes within company networks. the manufacturing processes, also to enhance planning accuracy and achieve cost reductions. In Smart Logistics, the implementation of sensors, Decision support systems can be used to allow a actors, and Cyber-Physical Systems (CPS) systematic evaluation among different scenarios in increases the accuracy and real-time availability of order to take better decisions. Moreover, Industry 4.0 information and lowers the costs of supply-chain-wide strategies can be seen as enablers of automation by tracking and tracing systems [13]. Also, monitoring of reducing the time from failure occurrence to failure

14 ICPC 2019 – 1.2 notification by automatically triggering fault-repair (RFID) technologies. Thereby, the IoT is used to actions through Smart Devices. The material flow can identify and track objects (e.g., products, container, be supported by digital assistance systems based on machines, vehicles) in logistics systems and supply augmented reality, employees get individualized chains. The objects are constantly processing information about necessary tasks to get along in information about their surrounding environment and timed productions and decentralized working stations can be unambiguously allocated which increases the could negotiate cycle times and thus find the optimum effectiveness and efficiency of all related monitoring between highest possible capacity utilization per and control processes [17]. working station and a continuous flow of goods [15]. Physical Internet (PI) TECHNOLOGY ENABLERS FOR SMART The PI is an open, standardized, worldwide freight SUSTAINABLE SYSTEMS transport system based on physical, digital and operative interconnectivity by using protocols, In this section, the authors outline a set of interfaces and modularization. A provider-free, technological concepts for Smart Logistics and industry-neutral and border-free standardization is discuss potential applications. Literature has one of the basic requirement for the PI which developed a multitude of divergent frameworks, connects and virtualizes material flows, in analogy to models and conceptualizations. Bechtold et al. (2014) the concept of the digital internet. Moreover, identify the following technology concepts as the main standardized containers and carriers are used to enablers of Industry 4.0: 1) cloud computing, 2) ensure a maximum utilization of transport vehicles mobile technologies, 3) robotics, 4) advanced and a better usage of spare capacities. These analytics, 5) machine-to-machine-communication, 6) principles can be applied in internal logistics systems, social media, and 7) 3D printing [11]. This as well as in transportation networks by using self- classification can be regarded as unspecific because controlling, autonomous systems in transport and of a missing unambiguous classification of storage processes as one of the central elements of technologies as enablers for Smart Production and/or the PI. The usage of shared transport capacities, Smart Logistics initiatives. Subsequently, the authors storage locations, hubs and delivery points will have describe the concepts of the Cyber-Physical Systems a positive effect on both economic (e.g., short (CPS), the Internet of Things (IoT) and the Physical transportation times, lower costs of human resources) Internet (PI) as the core elements of the proposed and ecological (e.g., reduction of traffic and Smart Logistics component within the Smart Logistics emissions) effects [18]. Framework. APPROACHES TO A MORE Cyber-Physical Systems (CPS) SUSTAINABLE DESIGN CPS are physical objects or structures, such as products, devices, buildings, means of transport, production facilities, logistics components, that include embedded systems in order to ensure interactive communication [16]. The systems are connected through local and global digital networks. CPS detect, analyze and capture their surrounding environment, using sensors data combined with available information and services. Moreover, actors are used to interact with physical objects. CPS act autonomously, decentrally, can build up network amongst themselves and can independently optimize themselves according to the principles of self-similar fractal production systems. The Smart Factory interacts with human resources and/or machines and is able to organize itself in a decentralized, real-time manner [16]. A virtual image of the reality (“Digital Twin”) is permanently analyzed and updated with Solutions for sustainability [20] real-time information, and continuously synchronized with information from the real environment. In this section, possible ways to a sustainable development in its economic, social and Internet of Things (IoT) environmental dimensions will be outlined. New tools and policy approaches will be required to design, The IoT as an essential part of CPS is commonly assess and implement programs that meet the associated with Radio-Frequency Identification

15 ICPC 2019 – 1.2 defined goals. We must be aware that humans are mechanical vs. organic, empiric vs.noumenal. This not good at giving up luxury goods and services. had negative impacts on our relationships with nature [19]. The present trade off sustainability model with its The anthropocentric, globalized culture of the past associated infrastructure (coexisting social, technical created dichotomic systems: artificial vs. natural, and natural infrastructures) is not sufficient to solve or to correct our global problems (Figure 3).

Model of sustainability and associated infrastructures [19]

Sustainability requires innovation. More is needed, 3. Effectice control of flows: than just repair damages done to the natural Advanced algorithms for scheduling and routing in environment. Figure 4 shows a model of maturity combination with learning systems can stages in the design of sustainable systems, significantly improve the performance and processes and products. throughput of transport networks. Flow objects are goods, containers, verhicles or humans. In general, the following approaches can be applied to design for sustainability: To name some examples and practices, autonous vehicles and car-sharing in combination will 1. Reducing the consumption of resources: significantly change the picture and mode of city This can be applied at the level of individuals, with traffic. regard to one’s personal requirements, but also in the professional sectors of industry, retail and Researchers at the Massachussets Institute of services, in product definition, in manufacturing Technology (MIT) Senseable City Lab, the Swiss systems, in transport networks and other systems. Institute of Technology (ETHZ), and the Italian National Research Council (CNR) have developed 2. Sharing objects: slot-based intersections that could replace traditional By jointly using resources, on a personal/private traffic lights, which in turn leads to a significant or enterprise level, utilization of devices, tools, reduction of queues and delays. This model is based machines, vehicles, or space can be improved, on a scenario where sensor-laden vehicles pass reducing the number or volume of total required through intersections by communicating and resources. remaining at a safe distance from each other, rather than grinding to a halt at traffic lights [21]. The shift from ownership to service use, as a major element of sustainable use, has become With car-sharing, the utilization of passenger cars available in private vehicle mobility. In order to would potentially increase from 2 to 80 percent, make service use a new lifestyle, policy according to a survey by the Hungarian Academy of instruments have to focus it directly, rather than Sciences. Since the vehicles are always on the move, just changing marginal economic costs. Networks hardly any parking spaces would be required, A study and procedures for pooling solutions have to be by the University of found that 93 percent of established. the parking space could be saved [8].

16 ICPC 2019 – 1.2

CONCLUSION The consistent traceability of material flows which is based on new methods of automated identification As a finding from the theoretical and conceptual and tracking, the development of self-controlled analyses, Smart Logistics will be able to contribute to autonomous systems for transport and storage and an enhanced efficiency of supply chains and to the increase usage of advanced data analytics can be sustainable growth of industrial enterprises based on regarded as the first steps on the path to Smart new business models by taking advantage of the Logistics. following potentials: In this context, the following future challenges are • Enhanced control of processes based on real- identified: The establishment of the provider-free, time information industry-neutral and border-free standardization of systems and interfaces in material- and information- • Dynamic and situationally-orientated design of flow processes, the definition of business cases in processes in the sense of adaptive and self- order to evaluate the benefits of investments in new controlling systems technologies, the development of innovative organizational models for the integration of new technologies, the assurance of the security for human • Utilization of synergies by sharing capacities on resources and data (Cyber Security), the further neutral and independent platforms development and promotion of digitalization competences in vocational education and training • Increased decision-making processes based on incentives and the continuing integration of emerging extensive data analyses (Data Analytics) in technologies (e.g., Artificial intelligence (AI), Machine combination with closed loop learning systems Learning, etc.) into logistics processes.

• Ability of a flexible and customized adaptation of REFERENCES products, services and processes [1] Gleissner, H., Femerling, J. Ch. 2013. Logistical • Individualization of designs, configuration Infrastructure, in: Logistics - Basics - Exercises - options, orders, planning procedures, production Case Studies, Springer 2013 processes and ongoing operation under economic conditions [2] United Nations 2015. Integrating the three dimensions of sustainable development: A • Effective man and machine interaction by framework and toolsUnited Nations publication, including new principles of work design and ST/ESCAP/2737 competence utilization [3] Fuglestvet, J., Berntsen, T., Myhre, G., Rypdal, K., and Skeie, R.B. 2007. Climate forcing from the • Realizing new potentials for logistics transport sectors, Center for International Climate management through the development of new and Environmental Research, business models and innovative services. https://www.pnas.org/content/pnas/105/2/454.ful l.pdf [accessed 08 May 2019] In general, based on the Industry 4.0 approaches, [4] Division for Sustainable Development Goals Smart Logistics can be identified as a crucial element (DSDG) in the United Nations Department of of digitalization in the industrial value chain. Smart Economic and Social Affairs (UNDESA) 2015. Logistics is based on the usage of agile cooperation Sustainable Development Goals, networks and on the information- respectively https://sustainabledevelopment.un.org/?menu=1 organization-based connectivity in order to enable 300 [accessed 08 May 2019] intelligent and lean supply chains. [5] United Nations Environment Management Group (EMG) 2019. Sustainable Infrastructure for the CPS, the IoT and the PI can be seen as integral SDGs, https://unemg.org/sustainable- concepts of Smart Logistics, aiming at an increased infrastructure-for-the-sdgs/ [accessed 08 May supply chain efficiency by the creation of (partly) 2019] autonomous systems and processes. Independent [6] Blackburn, A. 2017. SDG 9: Relevant UNECE platforms for logistics services (e.g., transportation, Work on Resilient Infrastructure. storage, packaging, control of material and http://www.unece.org/fileadmin/DAM/trans/main/ information flow) contribute to a better usage of SDGs/Workshop_1_-_October_2017/II_- resources and to the creation of new business _Relevant_UNECE_Work_on_Resilient_Infrastr models. Advanced data analytics provides tools for ucture_-_SDG_9__UNECE_.pdf [accessed 10 more efficient decision-making processes. May 2019]

17 ICPC 2019 – 1.2

[7] United Nations Economic Commission for [16] Bauernhansl, T., ten Hompel, M., Vogel-Heuser, Europe (UNECE) 2019. Transport and the B. 2014 (eds). Industrie 4.0 in Produktion, Sustainable Development Goals, Automatisierung und Logistik: Anwendung, http://www.unece.org/trans/transport-and-the- Technologien, Migration. Springer, Wiesbaden. sustainable-development-goals.html [accessed [17] Boyes, H., Hallaq, B., Cunningham, J., Watson, 10 May 2019] T. 2018. The industrial internet of things (IIoT): An [8] Pumhösel A. 2017. Weniger Autos in der Stadt analysis framework. Computers in Industry. der Zukunft. In: Der Standard, 21. Juli 2017, 101:1–12. doi:10.1016/j.compind.2018.04.015. https://derstandard.at/2000061580102/Weniger- [18] Montreuil, B. 2011. Toward a Physical Internet: Autos-in-der-Stadt-der-Zukunft [accessed 10 meeting the global logistics sustainability grand May 2019] challenge. Logistics Research. 3(2-3):71–87. [9] Kagermann, H., Helbig, J., Hellinger, A., doi:10.1007/s12159-011-0045-x. Wahlster, W. 2013: Deutschlands Zukunft als [19] Swiatek, L. 2019. From Industry 4.0 to Nature 4.0 Produktionsstandort sichern: Umsetzungsem- – Sustainable Infrastructure Evolution by Design, pfehlungen für das Zukunftsprojekt Industrie 4.0. in: Jerzy Charytonowicz, Christianne Falcão Abschlussbericht des Arbeitskreises Industrie Editors,Advances in Human Factors, Sustainable 4.0. acatech – Deutsche Akademie der Urban Planning and Infrastructure, Proceedings Technikwissenschaften e.V., of the AHFE 2018 International Conference on https://www.bmbf.de/files/Umsetzungsempfehlu Human Factors, Sustainable Urban Planning and ngen_Industrie4_0.pdf [accessed 15 January Infrastructure, July 21–25, 2018, Loews Sapphire 2015] Falls Resort at Universal Studios, Orlando, [10] Hermann, M., Pentek, T., Otto, B. 2016. Design Florida, USA Principles for Industrie 4.0 Scenarios. In: Bui, [20] Tischer, U., Charter, M. 2001. Sustainable T.X., Sprague, R.H., (eds). Proceedings of the Product Design, in: Sustainable Solutions, 49th Annual Hawaii International Conference on Sheffield: Greenleaf System Sciences: 5-8 January 2016, Kauai, [21] Tachet, R., Santi, P., Sobolevsky, S., Reyes- Hawaii. 2016 49th Hawaii International Castro, L., Frazzoli, E., Helbing, D., and Ratti, C. Conference on System Sciences (HICSS); 2016. Revisiting Street Intersections using Slot- 5/1/2016 - 8/1/2016; Koloa, HI, USA. Piscataway, Based Systems. PloS ONE, NJ: IEEE. p. 3928–3937. 2016http://dx.doi.org/10.1371/journal.pone.0149 [11] Bechtold, J., Lauenstein, C., Kern, A., Bernhofer, 607 L. 2014. Industrie 4.0 - Eine Einschätzung von

Capgemini Consulting: Der Blick über den Hype hinaus. Capgemini Consulting, accessed 2019 Jan 15. https://www.capgemini.com/de-de/wp- content/uploads/sites/5/2017/07/industrie-4.0- de.pdf. [12] Zsifkovits, H., Woschank, M. 2019. Smart Logistics – Technologiekonzepte und Potentiale. Berg- und Huettenmaennische Monatshefte. 615. doi:10.1007/s00501-018-0806-9. [13] Louw, L., Walker, M. 2018. Design and implementation of a low cost RFID track and trace system in a learning factory. Procedia Manufacturing. 23:255–260. doi:10.1016/ j.promfg.2018.04.026. [14] Qu, T., Lei, S.P., Wang, Z.Z., Nie, D.X., Chen, X., Huang, G.Q. 2016. IoT-based real-time production logistics synchronization system under smart cloud manufacturing. The International Journal of Advanced Manufacturing Technology. 84(1-4):147–164. doi:10.1007/s00170-015-7220-1. [15] Kolberg, D., Zuehlke, D. 2015. Lean Automation enabled by Industry 4.0 Technologies. IFAC- PapersOnLine. 48(3):1870–1875. doi:10.1016/j.ifacol.2015.06.359.

18

ICPC 2019 – 1.3 What is needed to ensure efficient and clean road transport?

Dorothee Saar Deutsche Umwelthilfe e.V.

Copyright © 2019 AVL List GmbH, Deutsche Umwelthilfe e.V. and SAE International

ABSTRACT The necessary switch to renewable fuels however requires significant decrease in energy consumption. Air quality and climate targets require additional efforts to reduce emission from road transport. Role of DUH Compliance with existing standards and the Deutsche Umwelthilfe (DUH) is an independent NGO definition of standards for future technologies are for environmental and consumer right protection necessary. Changes in technology – driven by founded in 1975. We are engaged for better nature stricter legal requirements and requirements from protection, transition towards renewable energy, customer side – can be a chance for the sector with better air quality and sustainable transport modes in regard to competiveness on the global market. But order to ensure a healthy environment for us and for also for new technologies, clear targets and future generations. For many years, we are engaged standards are needed. Different interests have to be for low emission and energy efficient transportation. considered - individual mobility, the economic Due to ongoing breach of binding EU air quality importance of the sector and urgent climate policy values in many German cities, we currently run 35 requirements. cases to achieve effective air quality plans. We also NGOs like Deutsche Umwelthilfe (DUH) stimulate a have been very active to elucidate the diesel scandal. broad public debate on real world emission from the In this context we also provide own on-road-emission sector and demand compliance with existing tests of more than 100 passenger cars [1] that also legislation. For new technologies, standards and include measurements of NOx-emission before and effective market surveillance have to be introduced. after mandatory software updates from various The current debate on a transition of transport also models. The result is disappointing, showing that includes aspects like the future role of public transport under cold ambient temperature emission might be as well as social aspects both from consumers and even higher than before. However, we also saw Euro from employees side. 6d temp vehicles performing with NO2 emission values way below the limit value (Figure 1). INTRODUCTION Figure 1 Euro 6d temp vehicle with low NO2 In Europe, about 80,000 premature deaths are emission values (source: DUH) contributable to high NO2 concentration [EEA 2018]. High NO2 concentration in urban areas mainly derive from (diesel-fueled) road transport. About 45% of traffic based monitoring stations show exceedance of NO2 concentration limit values in Germany in 2017 (similar in France, Italy, UK) [UBA 2019].

The commitments under the Paris agreement and several national and EU agreements require CO2 redution in the transport sector by around 40% until 2030. However, in Germany, CO2 emission from road transport are as high as in 1990 [UBA 2019]. Beside the increase of vehicles and motorization, the growing In the context of the diesel scandal, we started legal gap between official CO2 numbers and real world cases against authorities to get access to relevant consumption is responsible for that. information. This approach takes a very long breath

19 ICPC 2019 – 1.3 and both authorities and industry obviously have no does not detect malfunctioning exhaust cleaning interest in transparency. Our latest success is a final systems reliably and must thus be amended by decision against the Ministry of Transport to give emission tests for NOx under load and a proper insight in the protocols of the so called “Inquiry system to measure particle number PN for both diesel committee ” set up by the ministry. In the and gasoline. DUH together with manufacturers of course of the inquiry, the ministry commissioned measurement devices just recently presented emission tests on various vehicles. However, the data convenient technology and a proposal for a test has only partly been published. We expect to get full procedure [3] insight within this year. RDE for CO2 As a result of our legal cases for better air quality, Better control of real world emission is also necessary bans for specific diesel vehicles are discussed in a lot for real world CO2-emission. CO2 emissions are of German cities and are to implemented in some of currently measured under lab conditions only. them – among them Stuttgart, the home of Daimler, According to the independent ICCT, the gap between and Bosch. The administrative court just certification data and real world fuel consumption has announced recently that a ban for Euro 5 diesel has grown from 8% in average in 2001 to 39% in 2017 to be integrated in the local air quality plan by July under unchanged test conditions [ICCT 2018]. 2019. The new test procedure WLTP offers some However, taking legal action is only one part of our improvement towards more realistic conditions. work and we continue to seek the dialogue with all However, it still does not fully reflect driving on the relevant stakeholders and the public. Fortunately, the road and circumvention can still not be excluded. A diesel technology is not decisive for the compliance high WLTP as “starting point” for the next phase of with our climate goals. DUH has gathered some CO2 regulation for passenger cars as of 2021 would arguments to damask this persistent myth [2] weaken the effect of further CO2 mitigation since a percental reduction is required and not a concrete WHAT IS NEEDED TO COMPLY WITH value. Latest publication from Emission Analytics [EA THE COMMITMENTS? 2019] raises questions and support the demand of onroad emission testing in the certification procedure. European legislation on emission standards is We call for a methodology how to measure CO2 complex and rather ambitious. However, it lags emission on the road. First proposals are available. implementation, independent control and sanctioning Obviously, the procedure must be reproducible and which leads to emission of exhaust gases above the test conditions clearly defined. Independent third current limit values. parties proofing their competence should be included as well. Market surveillance and transparency Efficiency standards are necessary for vehicles The diesel scandal brought facts to light that were driven by renewable fuels as well. The fact that BEVs known to experts long before. Three and a half year are counted as “zero” with regard to their CO2 later, the problems still have not been solved. Despite emission and the current practice to assess plug-in- several requests based on freedom of information hybrids disguise the real world emission from those legislation, there is no disclosure of the assessment vehicles instead of incentivizing efficient models of defeat devices, requirements and effects of which are necessary to lower the total energy software updates – mandatory or voluntary. consumption in the sector. How should such Consumers still wait for adequate compensation and standards look like? Should they rather focus on the unrepaired vehicles move to Eastern European vehicle or on the battery? Should they also include countries, downgrading air quality on site. strategies for recycling and second life? The debate is open! The increase of regulatory details as well as complex technology and software features challenge market Only Technology? surveillance. A clear commitment to compliance with standards under “normal use” is thus what we call for. There is probably high technological potential for The basis for this requirement is set out in the EU further decrease of emission. However, this has to 715/2007 regulation. Independent third parties need stimulated by clear requirements and controlled. to be involved – at least until authorities and Existing legislation to lower CO2 emission from manufacturers provide their findings. passenger cars and trucks will contribute to the overall goal – namely if we do not manage to close Better market surveillance also means better the gap between real world emission and official periodical technical inspection. The current procedure numbers.

20 ICPC 2019 – 1.3

The ideas are not new. However, they are not • Forum Ökologische Marktwirtschaft FÖS 2018: implemented yet. Numerous studies recommend a Fair and low carbon vehicle taxation in Europe. change in taxation of fuels and vehicles that stimulate A report for Transport & Environment low emission vehicles and reflect the real costs of • ICCT: From laboratory to road: A 2018 update of burning fossil fuels. There are numerous good official and "real-world" fuel consumption and examples in Europe that show how targeted taxation CO2 values for passenger cars in Europe helps to lower average CO2 emission in the • Sachverständigenrat für Umweltfragen SRU passenger car fleet over the years [FÖS 2018]. 2017: Umsteuern erforderlich: Klimaschutz im Pricing CO2 even in the non-ETS sector transport Verkehrssektor. Sondergutachten 2017 needs to be tailored and has to take the specific • Umweltbundesamt UBA 2019: Pressemitteilung willingness to pay into account. A pricing will amend vom 02.04.2019 but not replace binding reduction targets. German • Umweltbundesamt: Handbuch government announced a climate law in 2019 to Emissionsfaktoren des Straßenverkehrs, define concrete steps that fulfil the 2030 target. A Version 3.3 good window though to come up with concrete solutions that not only will become effective in the far future. REFERENCES In addition, the way must be paved to push [1] (https://www.duh.de/projekte/eki-kontrollen/eki- sustainable transport modes that will also lead to a ergebnisse/) decrease in number of vehicles. More engagement is needed to promote and facilitate public transport and [2] https://www.duh.de/fileadmin/user_upload/downl also improve conditions for safe cycling and walking oad/Projektinformation/Verkehr/CO2- in urban areas. The connected transition in the sector Minderung/Infopapier_Diesel_Klimaschutz_M% needs to be discussed with the relevant stakeholders. C3%A4rz2018.pdf [3] https://www.duh.de/presse/pressemitteilungen/p CONCLUSION ressemitteilung/weiterentwicklung-der- periodischen-abgasuntersuchung-dringend- European legislation alone does not ensure the erforderlich-deutsche-umwelthilfe-ste/ transition needed in the transport sector but needs implementation and effective market surveillance. Subsidies on national level need to support sustainable technology and transport modes rather high emitting and oversized vehicles. A comprehensive legal and political frame is needed to stimulate a transition towards new transport modes and technologies that provides both short term and mid term decrease of harmful emission. This has to be supplemented by public debate about the future of mobility. Latest sales numbers might only mean a spot light. However, they see non-European car makers with e-vehicles on the top of the sales lists.

Clear guidelines and ambitious targets are also relevant to define the future role of this industry sector and its role in the global context.

LITERATURE

• Agora Verkehrswende 2019: Blog: Warum wir Regeln für die Effizienz von Elektrofahrzeugen brauchen • Emission Analytics 2019: The WLTP enigma • European Environment Agency 2019: Air quality report 2018

21

ICPC 2019 – 1.4 Economic Implications: Impact of e-mobility on supply chain and related market expectation

Gerhard Stempfer ZF Friedrichshafen AG

Copyright © 2019 AVL List GmbH, ZF Friedrichshafen AG and SAE International

ABSTRACT - Development of e-motors and inverters - Management of e-motor, inverter blocksets ZF is a global technology company and supplies systems for passenger cars, commercial vehicles, - Setup of modular software for HEV and EV and industrial technology, enabling the next How can these requirements and expectations be generation of mobility. With its comprehensive served at the necessary speed and within a wide technology portfolio, the company offers integrated spectrum of application-specific solutions? solutions for established vehicle manufacturers, mobility providers, and start-up companies in the Focusing on these targets, ZF decided to set up a fields of transportation and mobility. ZF continually new E-Mobility Project House that is characterized enhances its systems in the areas of digital by: connectivity and automation in order to allow vehicles to see, think, and act. - Serving truck-, bus-, off-highway-, and marine- In all of these applications, market and legislation related applications directly from one source with requirements have shifted to emission reduction or eMobility system solutions even locally defined Zero Emission standards - Working in explore mode and serving customer according to application-specific speeds and steps. demand with Minimum Viable Products (MVPs), Electrification-related technologies improve in partnership-based pilot projects at maximum significantly in parallel, and each respectively speed approaches “market availability” status. - Enabling application-specific system solutions Considering these changes, ZF prepared and is - Integrating all ZF competencies in e-motor and currently implementing the Next Generation Mobility inverter development and managing the e- strategy, in which electric mobility solutions are a key mobility solution blockset element for all aforementioned applications with a special focus on electromechanic drive and steering - Integrating internal and external industrialization systems. facilities and partners to prepare necessary global production capabilities The expectations of industry partners on this new With this Electric Mobility Project House, established drive and steering system is fostered by the speed of requirements and customer expectations will be market demand: Be fast in bringing out application- satisfied. specific system solutions without requirement specifications, but with system integration, Strengthened by this approach, ZF can support competence, and affordable costs. industry customers in implementing Next Generation Mobility and answer future societal demand with In addition to customer expectations, ZF’s strategy affordable emission-free or emission-reduced demands the following additional key competencies: vehicles. - System integration in HEV and EV

- Symbiosis of e-motor with transmission/axle

22 ICPC 2019 – 1.4

CHANGE OF REQUIREMENTS For truck applications, the expected scenario is globally more diverse and the electrification quote is In both society and industry-related applications significantly lower with later timing compared to relevant to ZF, requirements and market expectations passenger car applications, especially in the HD change quickly. segment.

Major changes include: The expected scenario for city buses is also globally very diverse, but increasing faster, whereas coaches • Increase of population more closely follow the truck evolution. • Urbanization • Local emission reduction Industrial applications are again significantly more • Decarbonization diverse with different scenarios. In marine • Digitalization technology, electric mobility will be more visible in pleasure craft than commercial fast craft. There is An increase in population causes an increase in high diversity in MD- and MD-material handling and resource and energy demands as well as emission special solutions for agricultural engineering, and levels. According to a number of estimates, two thirds again, very high diversity in construction equipment. of the world population will live in metropolises in 2050. Together, these two changes cause significant As an example, if we compare the expected scenario growth in urban areas and increase pressure to of e-mobility evolution construction equipment with reduce local emissions in metropolises. truck applications, we again see significantly lower values and later timing. We see global examples of low to ultralow emission zones and intense discussion of zero emission zones. Besides differences, it is also important to discuss Discussion about climate change is still controversial. common aspects within commercial vehicle and Scientists agree that the primary root cause of global industrial applications: warming is CO2 emissions caused by humans, methane, and other gases. Apart from differing a) Commonalities in performance and life cycle: opinions, decarbonization plans are underway in certain industries and regions. In addition to lower and, at least from the point of view As part of the industry, we need to develop of magnitude, comparable values in terms of power technologies that satisfy CO2 reduction plans. We and torque requirements, product life cycles are expect further reduction to lower values in additional comparable between commercial vehicles and applications, compared to passenger cars and trucks. industrial applications. The influences of digitalization lead in different b) Commonalities in e-mobility introduction: directions. On the one hand, e-commerce causes increasing goods transportation and will enable In nearly all commercial vehicle and industrial automated and autonomous vehicles to change applications, we expect market preparation or complete vehicle architectures. introduction of e-mobility solutions. All in all, we expect these requirement changes to result in disruptive technology changes in the vehicle c) Main market expectations: industry, including in electric drives, electric steering systems, and ePTOs. Based on these diverse e-mobility market preparation or introduction scenarios, the following expectations MARKET EXPECTATIONS RELATED TO are of high importance: E-MOBILITY • Providing e-mobility solutions starting at low The first aspect of e-mobility market expectation is and unpredictable volumes diversity related to markets, timing, applications, • Solutions enable same or increased and solutions. This means that in each application, performance level region or country the market introduction and • High durability from first product generation penetration scenario is different. • Very fast progress from first concept to SOP • System integration support necessary To handle this diversity, ZF works with scenarios in • Comparable low prototype costs different markets and applications. For example, in • Despite above expectations: attractive series passenger car applications, expectations are almost costs and outlook equal between conventional and electrified vehicles.

23 ICPC 2019 – 1.4

According to market expectations, an investment in g) Building blockset for electric motor and inverter new technology and solutions that is based on low and unpredictable volumes in parallel with attraction In 2017 and 2018, there was a program called IMS series costs is necessary. designed to develop a new electric mobility portfolio for industrial applications. The following In the following sections, we will discuss the related product positioning is the result of this program for ZF: ZF vision and selected strategic steps to satisfy these market expectations. - Electromechanic Steering Systems

NEXT GENERATION MOBILITY AND - Electromechanic Drivetrain Systems RELATED STRATEGIC STEPS - ePTO Systems Next Generation Mobility: These types of electromechanic systems, including Based on this requirement change and related market first pilot products, are and will be offered by ZF in the expectations, ZF prepared and rolled out its vision new portfolio, including e-motors, inverters, and called Next Generation Mobility, starting with the related controls. mission to offer clean and Zero Emission solutions. In order to be very fast in providing electromechanic This vision also applies to commercial vehicles and systems to the market, a strategic corporation was industrial applications, and is currently being set up with ZAPI Group. Thanks to combined forces implemented for these applications. and components, this corporation enabled early and very important common electromechanic systems, The electric mobility technology area states that for all which are already available on the market or being mentioned vehicle applications, ZF is developing and prepared for market launch within the next years. will develop and provide electric mobility system solutions, including related components and For the development teams at ZF in both commercial technologies. In addition to technology solutions, vehicle and industrial applications, e-motor and system integration is provided by ZF to implement inverter development as well as industrialization these solutions within the expected time line. competence and capacity was built up around two pillars: ZF will provide these solutions for the global market depending on demand and the focus of each region. a) Integrating technology experienced resources Additionally, we expect these new technologies and based within passenger car eMobility new market requirements to create new vehicle development concepts, especially by combining automated driving and e-mobility. b) Building up competence in specific product line teams Related ZF Strategic Steps: In order to implement this mission, many measures Additionally, within the program for IMS system and strategic steps are necessary, some of which integration, competence and capacity for hybrid have already been implemented and others planned. and electric vehicles was established and In the following section, we will discuss some implemented in the first pilot projects. selected strategic steps and their status within ZF: As electric mobility market expectations are very a) IMS electric drives and ZF e-mobility portfolio dynamic, the time to market is shorter compared to established technologies. In order to satisfy this b) Strategic partnerships expectation, ZF decided to work in different project steering and operation modes, enabling shorter time c) Integration of electric motor and inverter to market with more agility. competence One measure to facilitate this in the best way possible d) IMS electric drives and system integration was the setting up of a new organizational unit, Electric Mobility Project House, at the beginning of 2019. Here, electric mobility projects for e) Speed and MVP commercial vehicle and industrial applications will be implemented to produce the best possible f) Electric Mobility Project House between framework conditions to leverage synergies. commercial vehicles and industrial technology

24 ICPC 2019 – 1.4

Within this project house, a commonly used building CONCLUSION blockset for the electric motor and inverter will be developed and managed, enabling integration of Currently ongoing requirement changes will intensely common components in different commercial vehicle influence vehicle technologies and solutions in all and industrial applications. applications. Even if global market penetration needs time for all applications, the impact on the supply SUMMARY OF IMPACT OF E-MOBILITY chain is already a reality. ON SUPPLY CHAIN This means that the supply chain is currently Based on these changes, expectations, and project reinventing itself for new e-mobility solutions and experience, the following summary regarding impact adapting to it in terms of organization. on supply chain and on ZF can be provided: ZF considers electric mobility in its Next Generation Electric mobility solutions will be introduced for Mobility vision as a key technology and is preparing nearly all relevant applications; timing and market future solutions and the necessary organization very penetration scenarios are very diverse. carefully and with a high degree of importance. a) New portfolio and positioning is necessary. b) Investment in new products is necessary. The challenge is the right balance between current and new products. c) A shift of capacity is necessary, including the buildup of new competencies in e-motor and inverter development. d) System integration and cooperation are important. e) Very fast market launch is expected; higher speed compared to established markets. f) Attractive costs are a key element for the product and business case as well as acceptance of electric mobility.

In preparation for this cost level, ZF decided to modify the organization and set up the Electric Mobility Project House between CV und industrial applications in order to leverage synergies more comprehensively:

- In general regarding e-mobility technology with all electric mobility applications

- Within CV and industrial applications regarding systems and components

- Within electromechanic systems regarding transmissions and axles

25

ICPC 2019 – 2.1 Innovative EAS technologies in development for on-road and off-road applications

Rolf Brück Continental Emitec GmbH

Copyright © 2019 AVL List GmbH, Continental Emitec GmbH and SAE International

ABSTRACT Reducing raw NOX-emissions by adjusting the performance map and/or implementing engine Similar to passenger cars also for on-road and off- heating measures are difficult because of the road heavy-duty vehicles the challenge on the associated increase in CO2 emissions. Typical emission legislation side is comparable. The current exhaust gas aftertreatment systems for commercial discussion on tightening of NOX-limits in the vehicles comprise the following: commercial vehicle sector in Europe as well as especially in the Unites States presents a new Diesel oxidation catalyst (DOC): challenge for the engine- and catalyst-manufacturer. Here, engine HC and CO emissions are oxidized and, Lowering of 90 % compared to todays limits (down to if necessary, any additional fuel is used for increasing 0,02 g/bhp-hr) requires NOX-reduction at all engine the temperature in the exhaust system. Nitrogen operating conditions. Especially cold start and load oxide in the engine is converted to NO2 and used for points with low exhaust temperatures demand high “passive” DPF regeneration. activity of the exhaust aftertreatment system, particularly the amount and preparation of the Diesel particulate filter (DPF): reductant. This system filters out the particulates. Regeneration In passenger car application well proven close takes place either passively with NO2 (formed via the coupled catalyst configuration is deemed to be a DOC) or through an active increase in temperature major step to reach the future limits. The use of via the additional injection of fuel. Particulate filters additional heating measures seems unavoidable to often have an oxidation coating comparable to that in ensure the perfect preparation of the reductant on the DOCs. one hand side and to reach the needed conversion efficiency of the SCR catalysts on the other hand. Reducing agent dosage (watery urea solution: AdBlue®, DEF): The introduction of close-coupled catalyst solutions for heavy-duty applications requests new solutions on Provision of ammonia for reducing nitrogen oxide the substrate and catalyst side, as well as on the levels in the downstream SCR catalyst. canning side. Nitrogen oxide reduction (SCR): SUMMARY Selective reduction of the nitrogen oxide takes place with the ammonia (NH3) formed from the watery urea The estimated tightening of NOX-limits in the solution. The potential breakthrough of excessive commercial vehicle – without any negative impact on ammonia is prevented by the ammonia slip catalyst. greenhouse gas (CO2) emissions – requires NOX reductions under all operating conditions. Cold start Today in most Heavy Duty Exhaust systems these and load points with low exhaust gas temperatures components are located in the muffler mounted either place increased requirements on the activity of the at the frame of Heavy Duty vehicles or in underfloor exhaust gas aftertreatment system and, in particular positions for medium duty vehicle. here, on the quantity and conditioning of the reducing agent. For heavy duty catalyst systems the total catalyst and filter volume can go up to about 80l, compared to about 4,5l for a passenger car exhaust system. This

26 ICPC 2019 – 2.1 demonstrates the huge amount of energy needed to heat-up a truck catalyst system.

With the more stringent emission legislation, the cold start portion of the tailpipe emissions becomes of major importance. Also the temperature level at low load during city driving is getting a challenge, because similar to passenger cars, RDE or In-Use tests also in city driving will become the future.

By that the thermal management of the catalyst system becomes of major importance.

Beside moving at least a part of the catalyst into a close coupled position, active heating measures will be needed.

Todays truck engine compartments / frame designs do not allow to put the complete catalyst system close coupled. But moving the DOC and the SCR injection to a close coupled position seems possible in most applications in a first step. A vertical mounted rectangular metal substrate between engine and frame is an innovative solution. Also the so called “Universal Decomposition Pipe” supports a close couples DEF injection with a 100% evaporation of the droplets. By that flex bellows behind the close- coupled system are not negatively influenced by deposits.

Regarding active heating measures particularly an electrical heated catalyst, which is already known from passenger car and bus retrofit can be used. In combination with a HC-Doser the electrically heated catalyst works as a “match head”. A heating power of 10-15kW can be achieved.

It could be demonstrated that the combination of close couples catalyst, electrically heated catalyst and HC-dosing in front, is a high efficient solution to heat-up the catalyst system of heavy duty trucks and reduce the emissions significantly.

27

ICPC 2019 – 2.2

Lowest CO2 Emissions Despite Ultra-Low NOx

Gernot Graf, Helmut Theißl, Klaus Hadl, Anton Arnberger AVL List GmbH

Copyright © 2019 AVL List GmbH and SAE International

ABSTRACT conversion rates not only in the HDDTC but also in the RMC and NTE conditions. In Europe Euro VI D

Upcoming ultra-low NOx emission legislations in the will be replaced by Euro VI E from September 2020 USA and Europe require the fulfillment of lowest onwards, where the cold start as well as PN emission limits. Therefore, a highly efficient exhaust measurement for in service conformity (ISC) testing aftertreatment system in combination with a suitable will be included. Recently the “Euro VII” discussion control architecture, low NOx engine out emissions has started. Although there are no specific targets or and rapid heat-up measures become mandatory. numbers published yet, an enhanced focus will be, Using a 2-stage SCR system not only allows high beside NOx and PN, put on N2O and ammonia DeNOx-conversion immediately after cold start but emissions, not only in legal test cycles but also during also gives additional benefits in terms of passive real driving. Also in China, a significant reduction of regeneration, PN filtration efficiency, DeNOx pollutant emissions is expected. balancing and diagnostic strategy. In order to ensure a rapid heat-up and keep warm of the exhaust Proposed ultra-low NOx standards in North America aftertreatment system, a variable valve train is an require a NOx reduction down to 0.02 g/bhp-hr NOx attractive option for future commercial engine tailpipe. Additionally, CARBs objective is to develop a concepts. Additionally, greenhouse gas standards new low load cycle (LLC) to represent real world and CO2 limits respectively, require diesel engines urban tractor and vocational vehicle operation [1]. with 50% brake thermal efficiency. However, EGR will Moreover, the implementation of in use-conformity, still be mandatory to ensure emission compliance possibly similar to Euro VI E, to monitor real driving under an extended range of environmental emissions using the moving average window method, conditions. is under discussion. Lengthening the useful life up to

Figure 1 Emission legislation commercial vehicle on-road - tentative scenario

INTRODUCTION 550.000 miles (class 4-7) and 1.000.000 miles (class 8) respectively, requires a highly robust and reliable Future worldwide emission standards not only require aftertreatment system. Although CARB has allowed the reduction of local pollutant emissions, to improve to maintain the current OBD limits as long as low NOx especially urban air quality, but also a significant standards are optional, a tightening of OBD limits is lowering of CO2 emissions in view of climate change. also expected as soon as the standards become Upcoming ultra-low NOx legislation in North America mandatory. N2O and CH4 limits impose additional requires the fulfillment of lowest emission limits. A boundaries to the engine and aftertreatment layout. NOx reduction of up to 90% from the current standard is proposed by the Californian Air Resource Board (CARB), see Figure 1 leading to highest DeNOx

28 ICPC 2019 – 2.2

CO2 LEGISLATION ULTRA-LOW NOX REQUIREMENTS

The mitigation of the ongoing climate change, largely To highlight the extraordinary requirements of the affected by anthropogenic carbon dioxide (CO2)- ultra-low NOx standards in combination with Phase 2 emission, is a central aspect of global environment greenhouse gas standards in the USA, Figure 3 politics. In the USA for example, the environmental shows a comparison of the current emission protection agency (EPA) indicates the portion of the legislation with future requirements for a class 8 truck.

Figure 2 Worldwide CO2 Legislation total CO2-emissions coming from the transport sector In order to comply with the emission standard of with 26.3% [2]. In this sector commercial vehicles 0.02g/bhp-hr NOx emissions, the light-off of the contribute with 22.5% and have the fastest growth aftertreatment system needs to be ensured within a rate with 75% from 1990 until 2014. In Europe the very short time after cold start in the US HDDTC, commercial vehicles contribute with 6% to the total using low NOx engine out emissions. Even by CO2-emissions [3]. Therefore, a further increase of assuming no NOx tailpipe emissions afterwards, the freight efficiency and the subsequent reduction of SCR needs to be heated up within 250s using a CO2-emissions is a crucial goal for the future.

Consequently, governments worldwide have agreed on standards for GHG, CO2 and fuel consumption limits. Figure 2 shows a worldwide overview of the CO2 legislation for commercial vehicles. The USA follow the strategy to separate vehicle and engine targets and Canada has released regulations which are similar to the USA. The EU will release vehicle limits (in g CO2/tkm) and no specific engine limits. Since 2019 the engine is included in the monitoring process together with the vehicle and powertrain 2 x components, which influence the vehicle CO2 Figure 3 Comparison of current CO and NO emission. The vehicle CO2 emission will be emission limits with future requirements determined by a simulation tool (VECTO). A similar in US for a class 8 truck approach is used in Japan. China also limits the vehicle fuel consumption, although using a different calibration with 1.5g/kWh NOx engine out, evaluation process, with the chassis dyno test being respectively within 50s using 5g/kWh (see Figure 4). a unique approach for commercial vehicles. Considering a NOx conversion of <100%, light-off of

29 ICPC 2019 – 2.2 the SCR is mandatory no later than 150s after cold especially for applications dealing with strict start using NOx engine out emissions <1.5g/kWh. packaging constraints, concept 2 might be a viable solution.

Figure 5 Engine and aftertreatment concepts targeting ultra-low NOx (both equipped with cooled/uncooled HP-EGR and exhaust flap)

In Figure 6 a comparison of using EEVO with and Figure 4 Required DeNOx light-off depending on without DOC, upstream of the close coupled SCR for different NOx engine out scenarios in US concept 1, is plotted. Due to the thermal inertia of the HDDTC cold cycle (assuming 100% DOC, the light-off of the SCR is shifted by more than DeNOx afterwards) 100s. Therefore, placing the SCR as a first component is recommended as long as desulfation ENGINE AND AFTERTREATMENT and HC-storage of the SCR can be handled. CONCEPTS FOR ULTRA LOW NOX

The most promising engine and aftertreatment concepts from AVL perspective are illustrated in Figure 5 [4], [5]. Concept 1 consists of a 2-stage SCR system in combination with a variable valve train with early exhaust valve opening (EEVO) for temperature management. Usage of a DOC upstream of the close coupled SCR is optional and mainly depends on the used catalyst technology for close coupled SCR, HC poisoning and its desulfation strategy. Utilization of a box type layout is recommended for the components Figure 6 Influence of DOC on temperature downstream of the close coupled SCR/ASC to keep upstream of ccSCR by usage of EEVO the backpressure increase on a moderate level and for keeping synergies from the current Euro VI N2O is another gaseous emission which will soon be aftertreatment architectures. Moreover, 2-stage urea limited. On the one hand, N2O formation occurs on dosing allows a smart balancing of the NOx the oxidizing coating of the DOC, by incomplete conversion, giving additional benefits in terms of reduction of NO2 due to hydrocarbons [7], [8]. Low HC passive regeneration, PN filtration efficiency and engine out emissions, as well as an optimized diagnostic strategy [6]. catalyst coating technology are mandatory to reduce this effect. On the other hand, N2O is also formed on Concept 2 consists of a SDPF, again with a second catalysts with SCR coating, again depending on the urea dosing after the particulate filter. Rapid heat-up used technology [9]. Lower NOx engine out emissions is supported by a heater with ~12kW net power. A automatically lead to a lower urea injection, therefore SDPF with a single stage urea dosing is less the reduction of laughing gas emissions is feasible by attractive due to limited potential for passive the adaption of engine out emissions. Moreover, the regeneration. This leads to frequent active DPF N2O emissions are subject to variation depending on regenerations and consequently to a more severe the NOx conversion, which is controlled by the NH3 aging of the system and a higher fuel consumption for filling level at the SCR catalyst. An exemplary tradeoff the end customer. Yet again, usage of 2-stage urea of a typical EPA2010 aftertreatment system in US dosing shows advantages similar to concept 1. HDDTC is shown in Figure 7. However, SDPF is currently not the main technology route for on-road commercial vehicles which leads to high development and validation effort. Nevertheless,

30 ICPC 2019 – 2.2

temperature performance of the system. Disadvantages in the low temperature performance can be compensated by the usage of Cu-SCR technology for the close coupled SCR or SDPF.

Furthermore, using a 2-stage SCR system not only allows high NOx-conversion immediately after cold start but also provides additional opportunities in terms of soot load control on the filter (see Figure 10). The actual soot load on the filter defines the distribution of the overall urea dosing amount Figure 7 Tradeoff between NOx conversion and between the two SCR-stages and hence the NOx N2O tailpipe for 2 different NOx engine (NO2) to soot ratio upstream of the filter. In this way, out levels for a typical EPA2010 EAS the soot load can be balanced in a range that allows

Highest NOx conversion rates are not only required in a high PM Number filtration efficiency and low DPF- the HDDTC but also in the RMC and NTE conditions. backpressure [6], [10]. As mentioned above, the potential of a SDPF with a single stage urea dosing is seen as critical due to the limited NOx conversion at full load and non-standard conditions. Especially in aged conditions, the emission reduction at high temperatures is limited (see Figure 8Fehler! Verweisquelle konnte nicht

Figure 10 Smart DeNOx balancing to ensure optimum tradeoff between PN filtration efficiency and backpressure

Additionally, a controls and diagnostic strategy was developed by AVL for 2-stage SCR systems. Further Figure 8 Measured NOx conversion efficiency for a details are already described in the publications [6] Copper SDPF in fresh and aged status and [10]. gefunden werden.). EFFICIENCY GOAL FOR THE ENGINE To compensate the weak points and use the advantages of the different SCR technologies, a Figure 11 shows the required fuel consumption combination of Fe-SCR and Cu-SCR is targets derived from the US GHG Phase 1 and recommended. For the second SCR system in Phase2 legislation for heavy-duty diesel, which are concept 1 and 2 a Fe/Cu-SCR hybrid combination representative values for all developed markets. allows a compromise between the low and high temperature performance of both formulations (see Figure 9). Usage of a small Fe-SCR portion upstream of the Cu-SCR significantly increases the high

Figure 11 Fuel consumption targets for heavy-duty Figure 9 Advantages of Fe/Cu-Hybrid SCR vehicles in the USA combination 31 ICPC 2019 – 2.2

Figure 12 Building blocks for a high BTE

In order to achieve MY’27 CO2 limits, a cycle BSFC available turbocharging efficiency, as well as the value of 181 g/kWh in RMC is required. Considering desired EGR rate and EGR concept. The EGR a margin, a minimum BSFC of 170 g/kWh is concept and rate dictate the necessary pressure estimated, which corresponds to 50% brake thermal difference between the intake and exhaust manifold. efficiency (BTE). A further reduction of friction losses of the base BUILDING BLOCKS engine is impeded by the demand for a higher PFP capability. Nevertheless, measures on the base In general, a high BTE of an ICE can be realized by engine, as well as minimized parasitic losses optimization of the indicated efficiency and reduction (auxiliaries on demand) will result in a further of friction losses. Additionally, WHR systems can reduction of friction losses of future commercial exploit rejected heat, for example from exhaust gas engines. All these building blocks are summarized in and EGR mass flow. Figure 12.

The indicated efficiency is a combination of the high- pressure and low-pressure cycle efficiency.

The efficiency of the high-pressure cycle is influenced by the compression ratio, the cylinder charge mass, the combustion process itself and the wall heat losses. All these parameters affect the necessary peak firing pressure (PFP) capability of the engine. Therefore, an increased PFP capability is one of the main prerequisites for achieving a high BTE. Parameters for improvement of the low-pressure cycle (essentially pumping work) are a high volumetric and high charging efficiency, as well as lowest pressure losses in the intake and especially in the exhaust system.

The balanced optimization of both, high-pressure and low-pressure cycle is strongly influenced by the

32 ICPC 2019 – 2.2

Figure 15 BSFC - NOx trade off; w/ & w/o EGR

EXHAUST GAS RECIRCULATION

The measures for meeting the required CO2 and fuel consumption targets affect the boundary conditions for the aftertreatment system, such as exhaust gas temperature and NOx engine out level. Figure 13 Roadmap for CO2 reduction [5] The main factors to improve the gas exchange are the CO2 REDUCTION ROADMAP charging and EGR system. Therefore, one of the main focuses is the increase of turbocharger Figure 13 shows a reasonable scenario in view of efficiencies to reduce gas exchange losses. State of achieving MY’27 CO2 limits. The SET (RMC) related the art charging systems have an efficiency of 50 to CO2 emissions will be realized by using ”engine 55%. Future charging systems are expected to have internal” measures only. These measures are the efficiencies in the range of 60% and higher. improvement of combustion, gas exchange and friction as well as the adaptation of operation strategy. With such improved charging system efficiencies, it In this case, an average improvement potential has becomes increasingly challenging to generate the been considered. However, additional systems such required EGR rates for NOx reduction, with high- as WHR or hybridization might be implemented to pressure EGR systems only. Currently discussed meet 2027 legislative requirements. [5] counter-actions will require alternative EGR systems, such as EGR pumps or low-pressure EGR systems, PEAK FIRING PRESSURE additionally to (existing) high-pressure systems.

As a consequence of “engine internal” measures for Figure 15 shows the NOx BSFC trade off measured CO2 reduction, the PFP capabilities of the base on a R&D HD diesel engine at the load point 1200 engine need to be increased to levels in the rpm full load, in which the boost pressure and exhaust magnitude of 280 bar (see Figure 14). pressure level was kept constant. The blue curve represents a variation of injection timing without EGR. Obviously the minimum BSFC without EGR is reached at a high NOx engine out level. EGR is an enabler to realize the combination of low BSFC at low NOx emission. The red line represents an EGR sweep at optimal injection timing. In addition, the EGR technology provides a flexibility in NOx engine out in order to react to different EAS boundary conditions. Therefore, EGR becomes mandatory for future engine concepts.

Figure 14 Peak Firing Pressure increase

33 ICPC 2019 – 2.2

EXHAUST GAS TEMPERATURE Stoichiometric approach Compared to a diesel engine, the overall efficiency of Figure 16 shows the tradeoff between BSFC and the stoichiometric gas engine is lower. In its sweet cylinder mass ratio (CMR, detailed description in spot a stoichiometric commercial gas engine can publication [11]) for different TC efficiencies at an reach 40% BTE which results in 13% lower CO2 engine operation point of 1200 rpm - 25% load. The emissions compared to a HD diesel reaching 46.5% higher the TC efficiency, the higher the CMR for an BTE. However, the benefits are mitigated in the optimal balance between high pressure and low- partial load due to throttling losses and lower cylinder pressure cycle. A high TC efficiency is therefore not mass. Nevertheless, in a low load cycle such as the only a prerequisite for the reduction of pumping WHTC, a CO2 benefit remains. losses, but also for an improved high-pressure cycle.

Figure 17 Exhaust gas temperature decrease by increased cylinder mass Figure 16 Trade-off BSFC – Cylinder mass ratio at On the one hand, the low efficiency in the partial load an engine operation point of 1200 rpm – is a weakness of the stoichiometric engine. On the 25% load other hand, the usage of a three-way catalyst offers a great potential for pollutant emission reduction. CARB The trend towards higher cylinder mass, either driven recently published that all 10 engines, certified to the by the high TC efficiency or by the necessity of a 0.02 g/bhp-hr optional NOx standard in California, are reasonable EGR rate, reduces the exhaust gas natural gas or liquefied petroleum gas powered temperature downstream of the turbine (see Figure engines. [12] 17). Therefore, the conflict between a high BTE, reasonable NOx engine out emissions and a sufficient High pressure direct injection exhaust gas temperature for the EAS becomes progressively challenging. HPDI means high pressure direct injection of natural gas. The gas burns on a diffusion flame. A diesel pilot GAS ENGINE AS ENABLER FOR CO2 provides ignitable conditions for the natural gas in the combustion chamber. AND NOX REDUCTION Natural gas therefore burns on a similar excess air Due to its chemical composition, natural gas offers a ratio as diesel. Compared to premixed lean burn theoretical CO2 reduction potential of 25% compared concepts, HPDI can operate significantly leaner. to diesel fuel, assuming the same fuel energy Hence the diffusion combustion of gas can avoid provided. Additionally, natural gas also offers a great BMEP limitations due to knocking issues and avoid reduction potential for pollutant emissions. CH4 slip from flame quenching, crevice losses and scavenging. Gas engine technologies for commercial engines are available on the market. The two most promising All in all, HPDI combines the benefits of a diesel technologies for mastering future requirements, combustion process with the beneficial chemical regarding pollutant criteria and GHG emissions compositions of natural gas. reduction, are the stoichiometric approach and high- pressure direct injection (HPDI). Project HDGAS 2020 AVL, funded by the EU, demonstrated more than 22% CO2 reduction by high pressure direct injection of natural gas on a heavy- duty engine with 200 bar PFP capability. [13] In case

34 ICPC 2019 – 2.2 of a maximum PFP of 200 bar, a 300 bar injection [6] M. De Monte, S. Mannsberger and H. Noll, "SCR pressure was sufficient. As described previously the control strategies with multiple reduction devices PFP of the future diesel engine will rise, hence also for lowest NOx emissions," in Heavy-Duty Diesel the injection pressure of HPDI will have to increase, if Emissions Control Symposium, Gothenburg, HPDI is to maintain similar power density and 2018. efficiency as diesel engines. [7] A. Beichtbuchner, L. Bürgler, H. Wancura, M. Weißbäck, J. Pramhas and E. Schutting, "HSDI CONCLUSION Diesel on the way to SULEV - Concept Evaluation," in 21st Aachen Colloquium In order to comply with future ultra-low emission Automobile and Engine Technology, Aachen, legislations, the modification of current aftertreatment 2012. architecture becomes mandatory. The focus will be to [8] A. Beichtbuchner, L. Bürgler, E. Schutting, K. achieve high NOx conversion in the cold part of the Hadl and H. Eichelseder, "Diesel- legal cycle as fast as possible by a combination of Abgasnachbehandlungskonzepte für die engine and aftertreatment measures. AVL sees the Richtlinie LEVIII SULEV," in Internationaler combination of thermal management by a variable Motorenkongress, Baden-Baden, 2015. valve train with early exhaust valve opening, together [9] A. Newman and J. Matthey, "High Performance with a close coupled SCR system as the most Heavy Duty Catalysts for Global Challenges promising technology route. A possible alternative is beyond 2020," in Heavy-Duty Diesel Emissions a concept including SDPF with 2 stage urea dosing Control Symposium, Gothenburg, 2018. and a heater for thermal management purpose. [10] B. Brier, M. Tandl, A. Grauenfels and G. Graf,

In order to fulfill current CO2 requirements, a HD "Impact of particle number limitations on engine diesel engine has to reach a minimum BSFC in the and exhaust aftertreatment layout," in 10th range of 180g/kWh, which corresponds to a BTE of INternational Exhaust Gas and Particulate 46.5%. From today’s perspective 50% BTE is feasible Emissions Forum, Ludwigsburg, 2018. using only “engine internal” measures. Enablers are [11] H. Theissl, H. Seitz, G. Gradwohl and T. Sams, an advanced combustion and an improved air & EGR "Die Aufladung als Schlüssel zur weiteren path. As a consequence, the peak firing pressure Verbrauchsreduktion am modernen levels will be increased, and the exhaust gas Nutzfahrzeug-Dieselmotor," in 19. temperatures will be reduced. Therefore, a new Aufladetechnische Konferenz, Dresden, 2014. engine design requires a high PFP capability, as well [12] S. o. t. M. S. Control, "CARB Staff Current as effective heat up and efficient keep warm exhaust Assessment of the Technical Feasibility of Lower gas temperature measures. NOx Standards and Associated Test Procedures for 2022 and Subsequent Model Year Medium- REFERENCES Duty and Heavy-Duty Diesel Engines," CARB, California, 2019. [1] California Air Resources Board, "Heavy-Duty [13] A. Arnberger and A. Prochart, "HDGAS D7.1 - Low Nox," California Air Resources Board, 22 Assessment results of independent testing," EU January 2019. [Online]. Available: Comission, 2018. https://www.arb.ca.gov/msprog/hdlownox/hdlow

nox.htm. [Accessed 16 April 2019]. [2] Environmental Protection Agency, "Greenhouse Gas Emissions and Fuel Efficiency Standards for Medium- and Heavy-Duty Engines and Vehicles—Phase 2," 2016. [3] European Comission, "Comission Staff Working Document Impact Assessement," European Comission, Brussels, 2018. [4] M. Decker, H. Theißl, J. Schubert and W. Schöffman, "The Commercial Vehicle Engine of the Future Considering Emission Legislation," MTZ Worldwide, no. 11, pp. 26-31, 2017. [5] L. Walter, T. Wagner, H. Theissl, S. Flitsch and G. Hasenbichler, "Impact of CO2 and ultra-low NOx legislation on commercial vehicle base engine," in 4th International Engine Congress 2017, Baden-Baden, 2017.

35

ICPC 2019 – 2.3 Development and testing of an innovative gas engine for heavy duty applications

Stefano Golini, David D’amato, Sergio Giordana, Paolo, Grosso, Diego Iudice FPT Industrial

Anton Arnberger, Gernot Hasenbichler AVL List GmbH

Davide PAREDI Politecnico di Milano

Peter Grabner TUG

Copyright © 2019 FPT Industrial, AVL List GmbH and SAE International

ABSTRACT A revolution will take place in the next few decades, which will bring about different vehicles and fuels to The need to drastically reduce GHG emission in the move goods across the world. next decade will deeply change the technological solutions employed for long haul transportation. Electrical vehicles seem to be the most probable Several alternatives are possible and, among them, candidate to replace ICEs in the near future. It must natural gas engines have proved a reliable and be underlined, though, that the battery efficiency efficient way to curb GHG emission. decreases as the vehicle size increases [1] [2]: this means that BEVs could be a viable options for LCV FPT Industrial is currently the European leader in the but it will be more difficult for them to replace ICEs in production of NG engines and joined the research LH applications. Furthermore, the technical principles project HDGAS (co-funded by the EU), aimed at the for the electrification of trucks are similar to those development of NG powertrains for the 2020s, to offer available for cars but the greater size and weight of NG engines with increased fuel efficiency. trucks and their more rugged operations substantially increase the barriers to batteries serving as a This paper describes the challenges encountered substitute for diesel engines [2]. during the development of the HDGAS engine and presents results of the simulations and testing It must be underlined that the impact on the performed during the project. environment of this technology is not negligible: the LCA of BEVs shows that they have a global effect INTRODUCTION similar to traditional vehicles [3]: this is highly dependent on the way electricity is produced [4] and with the current European mix, in which renewable At the end of 2018 the European Council introduced energy accounts only for 26% of the total [5], the limits on CO2 emissions from HD vehicles: compared positive effect of BEVs on global warming is very to the 2019 average, a reduction of 15% is foreseen limited [1]. In the current situation, BEVs can be in 2025, which will become 30% in 2030. effective in reducing the pollutants level in urban areas: this factor must not at all be underestimated Compared to the uniformity of today’s HD vehicles but will have little or no effect in reducing GHG (100% ICEVs, diesel engines > 99%), in 10 to 20 emission [3]. years the landscape will be significantly more varied, because it will be impossible to meet these limits In addition, batteries need raw materials, such as relying only on the development of the diesel engine. lithium and cobalt, whose production must be

36 ICPC 2019 – 2.3 increased substantially in order to meet the global tank of 1400 kWh, that should guarantee a 700-km demand of BEVs and there is a risk that large scale range, will then stay between 56000 and 84000 $. exploitation of these resources could lead to Costs are expected to fall to 15-30 $/kWh, though at significant environmental impacts [4]. What’s more, a slower pace compared to FCs [2]. the effect on water reserves must in no way be overlooked. The time necessary to build the infrastructure for the new vehicles, be they BEVs or FCVs, is a factor that Fuel cells are another candidate foreseen to replace must be fully accounted for, as more than few years the ICE, at least partially, on LH missions. An FCV is, will be necessary to put in operation the over 25000 essentially an electric vehicle using hydrogen stored charging points (20000 DC 150-500 kW and > 5000 in a pressurized tank and equipped with a fuel cell for DC >500 kW, along motorways) and the 1000 on-board power generation. Hydrogen is stored on hydrogen refueling stations (500 for compressed H2 vehicles in dedicated tanks at pressures of 35 MPa to and 500 for liquefied H2) that are estimated as 70 MPa: the energy density is much higher compared necessary by 2025-2030 to meet the EU standards to batteries, but hydrogen storage still needs four on CO2 [7]. times more space to achieve the same range as conventional diesel technology [2]. In the meanwhile, decarbonisation will have to be pursued reducing ICE’s GHG emission and this can Today, around 50% of hydrogen is produced from be accomplished reducing the carbon content of the natural gas through steam methane reforming, and fuel: natural gas (a blend of different hydrocarbons, one-third comes from the refining process of oil; the mainly methane) is the immediate replacement for rest is produced from either coal or electrolysis [2]. diesel fuel. From this, it follows that the ideal pathway for hydrogen production is electrolysis, using renewable To begin with, combustion of NG produces the lowest energy. Alternatively, biomethane and the use of CO2 quantity per unit of energy of all other carbon capture and storage also provide another way hydrocarbons [8]: due to the fact that NG has the to generate hydrogen with low life cycle GHG lowest C/H ratio and the highest energy content per emissions [2]. unit of mass, its combustion generates 58.5 gCO2/MJ versus 78 gCO2/MJ for diesel fuel, a reduction around Currently, there around 500 FCVs (mainly cars and 25%, assuming the same thermal efficiency. The buses) running across several demonstration projects higher efficiency of the diesel engine lessens this globally, but the interest in hydrogen and FCVs is amount, but it remains still significant: a recent study growing: up to 400 stations are planned to be [9] showed that, on a well-to-wheel basis, a HD NGV operating in Germany by 2023 while California has using CNG will produce 16% less CO2 than the set the goal of having 100 stations by 2020 and has corresponding diesel vehicle. developed funding programmes to achieve this target [2]. Nevertheless, it must be considered that the time Moreover, NG can be obtained totally from renewable needed to bring hydrogen-refuelling stations online is sources: the number of biomethane plants in the EU significant (California estimates it at two years [6]) rose from 180 in 2011 to 420 in 2015 [2] and and that refuelling is, in any case, a very complex biomethane is obtained from agricultural residuals or operation, especially when using 70 MPa. from landfills, so it is not in competition with food. If NG is obtained from renewable sources, GHG The high cost of the main components is another emission will be, on the average, 23% of GHG barrier that must be overcome. Current cost of FCs is emission with fossil NG [9]. around 1000$/kW and it will be necessary to wait for high-volume manufacturing of next generation of FCs Since NG technology has been around for the past to bring the cost somewhere between 60 and 200 few decades, the distribution network is well $/kW [2]. The tank is another very expensive item, established and it is continuously growing (Figure 1). ranging from 40 to 60 $/kWh (at 70 MPa): a storage

37 ICPC 2019 – 2.3

Figure 1 Infrastructure of “alternative fuels” for HDV [7]

NG HDVs have been available for the past twenty THE HDGAS PROJECT years and the most recent products, such as the IVECO Stralis NP460 (powered by FPT’s Cursor13 The HDGAS project started in May 2015 and lasted NG), can replace the correspondent diesel vehicle, 36 months. It was co-funded by the EU in the both in terms of performance and in terms of range framework of HORIZON2020 and it involved 20 (more than 1600 km with two LNG tanks [10]). partners from 9 different countries across the EU, Furthermore, the use of NG can lead to important from both the academic and the industrial world. savings on the fuel cost: LNG price in China in the past decade has been, on the average, 55% of diesel Its goal was to develop, demonstrate and optimise fuel price, on an energy equivalent basis and the advanced powertrain concepts for NG engines, to number of LNG lorries skyrocketed from 7000 in 2010 perform thereof integration into HD vehicles, and to to 132000 in 2015 [2]. But also in Europe, NG is confirm achievement of Euro VI emissions limits, and significantly cheaper than diesel fuel and can tip the in use compliance under real-world driving conditions, overall economic balance in favour of the LNG vehicle while reducing CO2 emissions at least 10% with [10]. respect to 2013 state-of-the-art engines.

The already significant advantages of NGVs in terms Three different ICEs were developed in the project of lower GHG emissions and lower operating cost can along with innovative after-treatment systems. be improved by improving the efficiency of NG Moreover, a new concept of LNG tank was pursued, engines. FPT Industrial has developed NG engines in order to increase vehicles’ range. Finally, technical since 1990s and strongly believes that this engine is requirements and international/European standards one of the possible solutions (certainly the most for LNG fuelling interfaces were drawn. readily available) to reduce the environmental impact of transportation. Project Organization In order to reach this target and to offer to its In order to achieve the above-mentioned goals, the Customers more efficient and economically viable HDGAS project was organised in 8 work packages products, FPT Industrial joined the HDGAS project, (WP), as shown in Figure 2Fehler! Verweisquelle aimed at developing innovative gas engines for the konnte nicht gefunden werden.: 2020s.

38 ICPC 2019 – 2.3

Figure 2 HDGAS structure

• WP1 (led by AVL and Uniresearch) to manage Prof. Dr. Theodor Sams, from AVL Graz, was the the Project Project Coordinator. Further information can be found • WP2 (led by Daimler) aimed at the development at the Project’s website (www.hdgas.eu). of advanced tank systems as well as standardization of fuelling process and interface In particular, WP4, led by FPT Industrial, was devoted • WP3 (led by Ricardo) aimed at the development to the development of an innovative positive-ignition and demonstration of a new generation of ATSs engine exclusively fuelled by natural gas, and the • WP4 (led by FPT Industrial) aimed at the integration of the engine in a long haul truck, along development of an innovative positive ignition with the new LNG tank developed in WP2 and the NG engine innovative ATS developed in WP3. • WP5 (led by MAN) aimed at the development of a dual-fuel, port-injected engine Tests aimed at demonstrating the improved fuel • WP6 (led by Volvo) aimed at the development of efficiency were performed at test bench. The engine a high-pressure gas injection engine was tested in two different configurations, • WP7 (led by AVL) to determine testing stoichiometric and lean burn, in order to evaluate procedures and overall results assessment potentialities in fuel consumption reduction of both but • WP8 (led by Uniresearch) to disseminate in this paper we will concentrate only on the Project’s results. stoichiometric version.

39 ICPC 2019 – 2.3

The target of -10% in GHG emissions and +10% in THE DESIGN PROCESS performance (torque/power) with respect to 2013 state-of-the-art engines was reached applying It was clear from the beginning that the design of the several new solutions to the HDGAS engine, such as: HDGAS engine would have been a complex task, due to the features which required a totally new cylinder • An innovative combustion system, with the head as well as other fundamental components. intake ports and the combustion chamber specifically designed for combustion in spark- Some constraints were defined: ignition engines, hence creating tumble rather than swirl • First of all, the new cylinder head would have to • A new fuel system, which makes it possible to be installed on the existing cylinder block, so the employ different injection strategies in order to number and the position of the bolts as well the improve air/fuel mixing and, thus, combustion number and position of oil and water passages efficiency, reducing pollutant emissions at the were fixed same time • Secondly, the external dimensions of the new • Cooled EGR, to increase fuel efficiency engine would have to be as close as possible to • High energy ignition system. With the Corona the Cursor13 NG engine, under development at ignition system, the air/fuel mixture is ignited in a the time larger volume compared to a standard spark • Finally, the vehicular interfaces of the new plug ignition system, so burn delay and burn engine would have to be the same (dimensions duration can be significantly reduced, leading to and position) of the aforementioned Cursor13 a swifter combustion and improved late burn NG behaviour, reducing the risk of knock • Variable Valve Timing, using a cam-phaser, on The main characteristics of the engine are listed both the intake and the exhaust. Hydraulic valve below. lash adjustment was also employed • Double overhead camshaft with inclined valves. Architecture 6-cylinder in line

The high level of new content of this WP required the Bore x Stroke 135 x 150 mm joint effort of several partners, each with a different Displacement 12.9 dm3 role and task: Valves per cylinder 4 • FPT Industrial designed the new engine Camshaft lay-out Double over-head components and procured and built the Valve drive Finger follower prototype multi-cylinder engines. FPT also

developed the stoichiometric configuration at As we will see, the final configuration of the engine test bench derived from a close cooperation between design and • AVL tested the stoichiometric version on a single 1D and CFD calculations. cylinder research engine • IVECO integrated the HDGAS engine in the The Combustion System demonstrator vehicle, along with the new LNG tank and ATS The design of the combustion chamber is based on • BorgWarner supplied the high-energy Corona the pent roof concept, which has been proven to be ignition system the best configuration for spark-ignition engines, from • Politecnico di Milano performed CFD simulations the efficiency of combustion point of view [12]. of the stoichiometric configuration to assist the design of the combustion system The design of the combustion chamber had to take • Ricardo performed 1D and CFD simulations of into account the necessity to install the fuel injector the lean burn version and developed the same and Corona ignitor, which are rather bulky objects: configuration at test bench the two components were arranged in a configuration • Technische Universität Graz performed 1D which allows good accessibility (for installation and simulations of the stoichiometric version to maintenance) while keeping them as close as determine the best configuration of the engine in possible to the centre of the combustion chamber, to terms of subcomponents (EGR lay out, improve air/fuel mixing and to facilitate the mixture’s turbomachting, camshaft profile,…). ignition.

The final configuration of the combustion chamber is shown in Figure 3.

40 ICPC 2019 – 2.3

• Direct injection system • Corona ignition system.

The experience gained by FPT in more than ten years of designing high-performance NG engines greatly helped with this cylinder head. Therefore, some components (both geometry and material of valves and valve seats) and the cylinder head material are the same successfully used in other Cursor NG engines. The cylinder head material is a cast iron variant, particularly suitable for high temperatures found in stoichiometric combustion.

Figure 3 Combustion chamber

The intake ports were designed in order to reach the best values of tumble ratio and discharge coefficient: the latter is a measure of the turbulence level (beneficial for combustion propagation) in the combustion chamber, while the former quantifies the mass flow through the ports and into the cylinder.

The proposed design was checked with CFD, in order to find suitable values for both parameters. The Figure 5 Cylinder head discharge coefficient thus found was then fed into the 1D model in order to verify that it guaranteed a correct FEA of the cylinder head, conducted according to global behaviour of the engine. FPT’s standard practice, showed that the component can successfully withstand the thermal and As tumble ratio and discharge coefficients have mechanical loads expected during its mission. opposite trend, the best trade-off was found after three iterations between design, CFD and 1D The cylinder overhead is a large aluminum simulations: the final configuration is shown in Figure component which has multiple functions: 4 (see also Figure 19). 1. It houses the camshafts and the HVA 2. It serves as the cylinder head cover 3. It has a central, oil-free to room where the upper parts of the ignitors and of the injectors, and their electrical connectors, are located. The injection rail is also located here 4. It has an integrated rear pocket to accommodate chain drive, cam phaser and variable force solenoid. The axial dimension of the camshaft journals has Figure 4 Final configuration of the intake ports been kept to a minimum, in order to reduce friction. Moreover, the use of HVA eliminates the need for The piston head was shaped in a lenticular way, in valve lash adjustment, thus reducing maintenance. order to reach the desired compression ratio and pockets are present on the piston ceiling to avoid interference with valves at TDC.

The Cylinder Head and Overhead The cylinder head is the core component of the C13 HDGAS, and it incorporates all the new solutions introduced in this engine, namely:

• Pent roof combustion chamber • Inclined valves Figure 6 Cylinder overhead

41 ICPC 2019 – 2.3

Gear Train and Back End The plenum is connected to the cylinder head by means of 12 straight ducts, designed in order to The back end of the engine was also deeply modified, reduce pressure losses to a minimum and to keep as this is the first Cursor engine to employ a double charge flow into the various cylinder as uniform as camshaft. Because of this, the gear train currently possible. used on the Cursor13 engine was revised and it is still used in the lower part, close to the crankshaft, while Exhaust Manifold and EGR Circuit a chain is used in the upper part to drive the two camshafts. The EGR circuit is used to recirculate exhaust gas to the intake of the engine in order to reduce gas The flywheel housing was completely redesigned to temperature during combustion, thus lowering NOx house the new components. production, wall-heat losses and sensitivity to knock; moreover, EGR reduces pumping losses in part-load operation [12].

As it is customary with 6-cylinder engines, the turbocharger has a twin scroll, to avoid interference between pressure pulsations of the various cylinders. Therefore, it was decided to extract the EGR from both branches of the exhaust manifold. For the same reason, the turbocharger is equipped with two WGs, one for each turbine scroll: this configuration improves the engine efficiency, keeping both sections of the engines (i.e. cylinders 1 to 3 and cylinders 4 to 6) in the same conditions Figure 7 Back end The EGR routes are separated down to the EGR Intake Manifold cooler and they are joined together only upstream the The intake manifold consists of two elements joined EGR control valve; reed valves are used to eliminate together: the fresh air-EGR mixer and the plenum. pressure fluctuations.

In the mixer, fresh air enters through the throttle valve, The FEA of the exhaust manifold showed positive coming from the charge air cooler, while EGR is results for this component too. added by the EGR nozzle, after it has passed through the cooler and its valve. The position of the EGR nozzle was determined with the aid of CFD calculation to obtain the optimal mixing between EGR and fresh air.

Figure 9 Exhaust manifold, TC and EGR cooler

External Packaging The external shape and dimensions of the Cursor13 HDGAS engine were kept as much as possible similar to the current Cursor13 engine. Most interfaces are in the same position as those of the Figure 8 Intake manifold and EGR circuit Cursor13 (intake duct upstream of the throttle valve,

42 ICPC 2019 – 2.3 turbine and compressor outlets) and the lay-out of the not exist, the model was calibrated and validated external components (turbocharger and exhaust against measured data of an existing FPT NG engine. manifold, EGR cooler and circuit) was designed in Later the model was adapted to the requirements of order to avoid interferences with the vehicle. the HDGAS engine (Figure 11).

The result is a compact and elegant engine which can EGR in Stoichiometric SI Natural Gas Engines easily fit in the IVECO Stralis frame. With the introduction of the EURO VI emission legislation, the operating mode of natural gas positive ignition engines was shifted from lean towards stoichiometric operation. Thereby, a three way catalyst can be used to comply with the very stringent emission limits. However, stoichiometric operation entails negative effects on the fuel efficiency and the thermal strain of the engine. EGR has received attention for the application in NG positive ignition engines to mitigate these penalties [14][15], due to its multiple effects.

Recirculated exhaust gas dilutes the fresh charge, increasing the inert mass in the combustion chamber and lowering the combustion temperature: as a consequence, also the exhaust gas temperature is reduced, and this lessens the thermal stress on several engine components, especially at the turbine Figure 10 Cursor13 HDGAS engine inlet, which is critical at full load operation. The reduction of combustion temperature leads to a THERMODYNAMIC ANALYSIS decrease of wall-heat losses and this increases fuel efficiency. In addition, cooled EGR reduces the knocking probability: thereby, the compression ratio of the engine can be raised and the combustion can be advanced to increase efficiency. Furthermore, the caloric properties of the charge during compression are altered: the isentropic exponent and the compression work is reduced, and fuel efficiency benefits from this.

Finally, the intake manifold pressure must be raised, when EGR is added, to maintain the load. In partial load operation this results in de-throttling and a reduction of the gas exchange losses. Also at full load operation the gas exchange losses can be reduced in combination with a waste gate turbocharger. The higher boost pressure leads to closing of the waste gate and, because of EGR, the turbocharger operates

in an area of better efficiency Figure 11 Simulation model of the HDGAS engine [13] A trade-off exists between the gas exchange losses and the achievable EGR rate, for a given EGR layout. The 1D engine simulations were performed at the A positive pressure gradient between exhaust beginning of the project to define the thermodynamic manifold and intake manifold is required to recirculate layout of the HDGAS engine. This included valve exhaust gas. The higher the amount of EGR, the timing, so different concepts for the gas exchange, higher the required pressure gradient and thus the like Early and Late Miller timing, were evaluated. higher the gas exchange losses. Several EGR Moreover, the turbocharger matching and the layouts were evaluated to minimize the pressure definition of the EGR layout were performed: for these losses over the EGR duct and thus limit the required purposes, a 1D simulation model was created with pressure gradient: Figure 12 shows the chosen AVL BOOST v2013.2 software. Model calibration is configuration which outperforms the other variants an essential part of 1D engine simulations and, since investigated and is the only one capable of getting at the time of the simulations the HDGAS engine did close to the EGR target curve, defined for each

43 ICPC 2019 – 2.3 engine speed at full load operation (curve d in Figure described effects results in a rise of the net indicated 13). efficiency, by 1.9%Pt. [13]. These results clearly demonstrate the positive effect of EGR on fuel efficiency and thermal stress of natural gas positive ignition engines and confirm the results published in literature.

Figure 12 The chosen EGR layout [13] Figure 14 The effects of EGR at partial load operation [13]

Early and Late Miller Valve Timing Apart from conventional valve timing, which is optimised for maximum volumetric efficiency, the Early and Late Miller (a.k.a. Atkinson) cycle were investigated: they are characterised by an early, and, respectively late intake valve closing (IVC) while, from a thermodynamic point of view, the processes are identical.

Compared to a conventional valve timing, the early IVC allows overexpansion of the fresh charge, which is then compressed again but the compression above intake conditions only starts when the piston again reaches the position it had at IVC: this reduces the volumetric efficiency, de-throttling the engine and Figure 13 Calculated EGR rate with different decreasing gas exchange work. layouts vs. target [13] Charge temperature undergoes the same changes, The Effects of EGR at Partial Load therefore it is lower, at a given crank angle, compared The simulations results with and without EGR are to standard valve timing: this beneficially affects the shown in Figure 14Fehler! Verweisquelle konnte conflict between knock, compression ratio and nicht gefunden werden. to quantify the effects of maximum attainable load [13]. EGR. The part load operation point at 1200 rpm and 8 bar BMEP is chosen for this comparison. In this condition the engine is throttled and, with EGR, the intake manifold pressure must be raised by 200 mbar in order to maintain the load. The exhaust manifold pressure remains almost unchanged, consequently the gas exchange losses decrease by 200 mbar. The reduction of the combustion temperature is observed in the wall heat losses (-13%) and the exhaust gas temperature (-120 degC). The reduction of the wall heat losses and the change in the caloric properties of the cylinder charge cumulate and increase the gross indicated efficiency by 1.2%Pt. The sum of all

44 ICPC 2019 – 2.3

Figure 15 Valve lift profiles Figure 16 The compressor operating map and the operating points at full load with different The described effects, valid for Early Miller timing, valve timings [13] also apply to Late Miller timing. The influence of Miller timing on the net indicated Figure 15 shows the intake and exhaust valve lift efficiency, which includes the positive effect on the profiles used for the simulations. gas exchange losses, is shown in Figure 17: “Early Miller” timing increases the efficiency by 0.3%Pt at The effects of early and late Miller in rated power 1900 rpm while with “Late Miller” profile the conditions (1900 rpm-370 kW) are shown in Figure improvement is 0.1%Pt [13]. 17: keeping the temperature at the intercooler outlet constant, the in-cylinder temperature at firing with “Early Miller” profile is between around 30 degC lower than with the standard profile, while with “Late Miller” profile the reduction is around 10 degC [13].

The intake manifold pressure must be raised by more than 500mbar with “Early Miller” timing and by 300mbar with “Late Miller” timing. Contrary to the EGR case, the exhaust manifold pressure also rises considerably, therefore, the reduction of the gas exchange losses is lower, always compared to the no Figure 17 Effects of different valve profiles at rated EGR vs.EGR case. power [13]

The reason is the turbocharger efficiency, as demonstrated in Figure 16Fehler! Verweisquelle konnte nicht gefunden werden.: the introduction of EGR shifts the operating points of the compressor towards higher efficiency, from the black curve to blue one, so that the compressor works in optimal conditions. With both Miller timings too the compressor works close to the optimum, but the difference, compared to the EGR case, is very small, and the efficiency remains approximately constant, because the maximum has already been reached [13].

45 ICPC 2019 – 2.3

Finally, Figure18 depicts the p-V and T-V diagrams: The work activity was divided into three steps: IVC is defined at 1 mm effective valve lift and added in the graphs. The overexpansion in the case of “Early 1. At first, steady-state flow bench simulations were Miller” timing, the higher intake manifold pressure and performed. Different configurations of the engine the lower charge temperature are well observed [13]. intake ports were tested to find the best compromise between engine tumble intensity and flow discharge coefficient 2. Different engine operating points and valve lift profiles, at partial and full-load conditions, were tested by means of transient cold-flow, full-cycle simulations. The evolution of in-cylinder tumble motion and turbulent kinetic energy were assessed 3. Finally, fuel injection was introduced and the efficiency of the air-fuel mixing process was evaluated taking into account the homogeneity level of the mixture inside the combustion chamber. One engine operating point at partial load was tested, as well as the full-load condition with two different intake valve lift profiles.

Steady-State Flow Bench Simulations Proper in-cylinder filling and charge flow motion intensity are fundamental to have a stable and efficient combustion process, so steady-state flow bench simulations focused on the evaluation of the Flow Coefficient (Cd) and the Tumble Ratio (TR). The former was expressed in a normalized form according to

where ṁ is the measured mass flow rate while ṁth is the maximum theoretical mass flow rate defined as

Figure 18 The p-V and T-V diagrams for Early

Miller, Late Miller and standard timing [13] where:

Z Number of valves CFD SIMULATIONS dv Valve inner seat diameter CFD simulations were carried out using the open p Pressure drop across the valve source OpenFOAM® technology coupled with the  Air density LibICE, which consists of a set of applications, solvers and libraries specifically developed during the years at the ICE Group of Politecnico di Milano. Such TR was defined as numerical methodology was extensively validated in previous works [16][17][18]. Simulations were performed using a Reynolds-averaged Navier-Stokes turbulence modelling approach. where 휔퐹퐾 and 휔푀표푡 are respectively the angular velocity of solid body rotation and the engine angular speed.

46 ICPC 2019 – 2.3

Full-Cycle Cold-Flow Simulations Once the geometry of the intake ports was defined, the second step was the simulation of the flow field in the cylinder with moving boundaries, during a whole working cycle, not taking into account fuel injection and combustion.

The aim of this simulation is the analysis of the flow motion evolution during intake and compression strokes in order to determine the conditions (mainly turbulence level, to guarantee a complete and fast combustion) in the proximity of the igniter at ignition timing.

The operating condition which were analysed are listed below.

Full load 370 kW @ 1900 rpm Full torque 2200 Nm @ 1000 rpm Partial load 100 kW @ 1200 rpm

Figure 19 Evolution of intake ports layout [19] Calculations were started at 335 deg CA with a timestep of 0.005 deg CA. Figure 19Fehler! Verweisquelle konnte nicht gefunden werden. reports the three different ports Boundary conditions derived from 1D analysis were layouts tested under steady-state flow bench imposed. The in-cylinder TR was evaluated according conditions. Layout (a) represents the initial geometry, to the methodology proposed in [20]. while (b) and (c) represent the modifications which were carried out on the basis of the results of CFD Valve lift profiles are shown in Figure 15. simulations. Layout (b) is characterized by a larger section aimed at enhancing the mass flow rate. Figure 21 displays the TR trend for the full-load Layout (c) displays intake ducts which are more operating condition with the “Early Miller”.profile. oriented towards the opposite cylinder wall, similar to geometry (a), but with an increase of the flow cross sectional area close to the caps of the valves stems.

Overall, Figure 20 demonstrates the capability of layout (c) to provide a good compromise between the discharge flow coefficient and the tumble motion intensity so this configuration was employed for further simulations and was used for the cylinder head casting.

Figure 21 TR evolution: full-load [19]

The shape of the curve is consistent with what was achieved in previous full-cycle simulations of DI engines [21], with a peak during the intake phase, a decrease and then a lower peak before the end of the compression phase.

Figure 20 TR and Cd for the different layouts of the It was then decided to compare the evolution of the intake ducts [19] turbulence level for different intake valve lift profiles.

47 ICPC 2019 – 2.3

Figure 22 shows the result for “Early Miller” vs. “Late Natural gas with a volume fraction of 84.7% in Miller” profile: it is quite evident that the “Late Miller” methane was injected through a centrally mounted, profile exhibits a higher level of turbulence (i.e. TKE) multi-hole, direct injector with a pressure of during the last phase of the compression stroke, and approximately 20 bar. this should guarantee a faster flame propagation, which will result in higher engine efficiency. The The different SOI timings used are listed below difference arises because, with the “Early Mller” profile, the lower valve lift combines with the early IVC 424 deg CA to destroy quite quickly the big-scale structures: as a 1300 rpm-2 bar 444 deg CA consequence, this leads to a turbulence decay in the first part of the compression stroke. With the other two Full load (both profiles) 413 deg CA profiles, and especially in the “Late Miller” case, the intake valves remain at the maximum lift (which is The computational mesh was refined to better considerably higher than in the “Early Miller” profile) capture the complex phenomena occurring during the for a long interval: the result is a macro-vortex which injection phase and its interaction with the air flow. remains basically unchanged during the piston descent and which is “compressed” when the piston The efficiency of the air-fuel mixing process was moves toward TDC, so that the macro-scale vortex is evaluated in terms of: converted to small-scale turbulence, as demonstrated by the second peak in Figure 22Fehler! • Relative air-fuel ratio Lambda, defined as Verweisquelle konnte nicht gefunden werden.. Contrary to what was found in the 1D analysis, the “Early Miller” timing seems to negatively affect the combustion process compared to the Standard or the “Late Miller” profile.

Since each case was simulated under stoichiometric conditions, Lambda was always expected to be unitary at the end of compression stroke. The stoichiometric air-fuel ratio of the natural gas composition used is 15.46.

• The Homogeneity Index (HI), which accounts for local distribution of the fuel (=1 in case of ideal homogeneous distribution)

The evolution of HI for the “1300 rpm-2 bar” case is reported in Figure 23Fehler! Verweisquelle konnte nicht gefunden werden.: the air-fuel homogeneity Figure 22 TKE evolution for two different valve lift gradually increases during the injection phase, almost profiles, full load : (a) “Early Miller”, (b) reaching a unitary value at the end of the “Late Miller” [19] compression stroke and this trend is consistent with what was observed in previous works [21]

Full-Cycle Simulations with Natural Gas Injection The influence of gas injection on the efficiency of the air-fuel mixing process was then evaluated.

Different engine operating conditions were considered and multiple valve lift profiles were used to verify the effects on mixing. The conditions simulated were:

1. 1300 rpm, 2 bar BMEP with “Early Miller” profile 2. Full-load with “Early Miller” profile 3. Full-load with “Late Miller” profile.

48 ICPC 2019 – 2.3

Figure 23 HI evolution for “1300rpm-2bar”: (a) Figure 25 Distribution of in-cylinder equivalence SOI=424 deg, (b) SOI=444 deg [19] ratio during the compression stroke; full load, “Late Miller” profile [19]. Figure 23 also demonstrates that there is very little difference between the two different SOI, even Finally, Figure 25 reports the distribution of in- though the case with earlier SOI shows a slightly cylinder equivalence ratio (i.e. 1/ lambda) for the full higher HI at the end of the compression stroke, load simulation with “Late Miller” profile on two because of the longer time available for mixing. orthogonal planes during the compression stroke. An almost homogeneous mixture (equivalence ratio Figure 24 shows the probability mass function (PMF) equal to 1) is found close to the ignition location, thus of Lambda, calculated for the “Late Miller” case at 720 increasing the efficiency of the ignition process, deg CA: as expected for a stoichiometric condition, though it is possible to discriminate lean zones, close most of the mixture is characterized by a lambda to the cylinder liner, and rich zones close to the piston: value close to one, thus ensuring a stable combustion the first ones might lead to HC formation while the process even in full load. second ones can lead to CO generation. However, the computed high levels of in-cylinder turbulence are expected to enhance the mixing during combustion

SINGLE-CYLINDER ENGINE TESTING

The single cylinder research engine was operated on an AVL test bed in Graz, Austria where the evaluation of different camshaft profiles, injector configurations, cooled EGR and the combustion chamber itself was performed.

The main advantages of single cylinder engines are lower costs for prototype hardware and lower assembly time for hardware change on the test bed. Since single cylinder engines are externally charged, Figure 24 In-cylinder lambda distribution at 720 deg they provide a degree of freedom in setting the CA; full load, “Late Miller” profile [19] operation points, like boost pressure and exhaust back pressure.

For the investigation of the stoichiometric concept, it brought the following particular benefits:

• Operation with rapid prototype engine control unit possible • One instead of six cylinders brings a cost reduction, since fewer cost-intensive prototype parts are necessary • Drift or malfunction on prototype parts are relatively easy to detect • The operation of the engine is independent from turbocharger hardware (freedom in boost pressure and EGR rate). This gives more liberty to investigate combustion parameters and its sensitivity to them • Parameters can be investigated independently. Lambda independent from gas exchange work, independent from MFB50.

Characteristics of the single cylinder research engine, based on the HDGAS engine (Figure 26) are:

• Bore x Stroke=135 x 150 mm • Swept volume= 2.1 dm3

49 ICPC 2019 – 2.3

• AVL Rapid Prototype Engine Management System for closed loop AFR control • Gas injector designed for high injection pressure • Cylinder head for single cylinder operation with innovative charge motion • Boost pressure control by an external supply • Back pressure control by an electrically-actuated back pressure valve

Figure 27 Influence of EGR on different parameters in partial load [22]

Figure 28 shows ignition timing variations for different EGR rates in partial load conditions: increase in the EGR rate leads to increase in brake thermal efficiency of the engine. For a constant lambda set point, more EGR reduces CO emission while CH4 increases.

Figure 26 AVL SCE, based on the HDGAS engine [22]

Potential of cooled EGR MFB50-EGR variations on a single cylinder engine can show whether desired EGR rates can be applied for multi-cylinder engine applications. The maximum EGR rate is usually limited by misfire, HC emissions or instability in combustion (COV).

Figure 27 shows the influence of MFB50 and EGR Figure 28 Influence of EGR on different parameters rate on brake thermal efficiency, delta pressure over in partial load [22] the engine, CH4 emissions and COV for the partial load point (1200 rpm-8 bar BMEP). Figure 28 also shows that combustion duration (measured as A_I90 minus A_I10) increases with the percentage of EGR. And cycle to cycle variation increases too while the rate of heat release is stretched (Figure 29 and Figure 30): nonetheless, these negative effects are outweighed by the positive ones: reduced wall-heat losses and reduced scavenging work. Furthermore, in full load conditions EGR can help reduce knocking, allowing advanced spark timing and improving fuel efficiency.

50 ICPC 2019 – 2.3

Figure 29 In-cylinder pressure and rate of heat Figure 31 In-cylinder pressure and rate of heat release for different EGR rates in partial release for different intake profiles in full load (average cycle) [22] torque (average cycle)

On the SCE, the effects of the different evolution of turbulence highlighted by CFD (Figure 22Fehler! Verweisquelle konnte nicht gefunden werden.) are clearly visible on the combustion process and offset the thermodynamic advantages shown by 1D analysis which, due to its intrinsic limitation, is not able to capture the flow evolution in the combustion chamber.

All in all, the thermodynamic benefit of a late IVC can be utilized without major changes of the combustion concept. Achieving similar thermodynamic benefits Figure 30 In-cylinder pressure and rate of heat with an early IVC requires a careful adaptation of release for different EGR rates in partial charge motion to avoid disadvantages in the load [22] combustion process.

Investigations of alternative camshaft profiles MULTI-CYLINDER ENGINE TESTING As we have seen, three different intake valve camshafts were foreseen during the SCE test The multi-cylinder prototype engine was assembled campaign. A volumetric efficiency optimized intake in proto workshop in the FPT’s plant in Bourbon Lancy valve timing, which was basically a carry-over from (France) and it was then delivered to FPT’s testing the standard diesel application, as well as two Miller facilities in Foggia (Italy), where it was installed on a camshafts: the design target for both Miller timings dynamic test bench. was the same volumetric efficiency at low engine revs, in order to guarantee the same low-end torque A complete overhaul of the control system had been (Figure 15). going on in parallel to the design and assembly of the prototype, due to the new content of the engine. The Figure 31 shows in-cylinder pressure and rate of heat new injection and ignition systems, the VVT and the release from SCE operation for the three different EGR valve, all required modification to the existing camshafts. control system: a new HW ECU was necessary to manage the new components and an auxiliary Smart Indeed, combustion with the “Early Miller” profile Driver Unit was introduced to control the injectors, results in a significantly slower turbulent flame which needed 65V to operate correctly. From the SW velocity and therefore longer heat release. Important point of view, new control strategies were created. for this comparison: all different valve lift curves were evaluated using the same intake port design. Experimental set-up The HDGAS engine was installed on a dynamic test bench, able to simulate also motoring conditions, featuring a 450 kW/3500 Nm dynamometer.

51 ICPC 2019 – 2.3

This test bed is normally used for development As mentioned in the CFD section, calculations purposes and it is certified according to Accredia; it is showed that, with the “Early Miller” profile, turbulence equipped with the following instrumentation: is less favourable, compared to the “Standard” profile, thus worsening engine’s performance: this was • ABB sensyflow FMT700-P to measure air flow confirmed during the tests with the SCE, even though • Micro motion CMF025 to measure fuel flow the difference was not so huge. • AVL AMAi60SII-R2C-EGR analyser to measure CO, THC, O2, CH4, CO2 It was decided to test the “Early Miller” profile also on • AVL LDD analyser to measure NO, N2O, H2O, the MCE, to further verify the results of CFD and SCE: NH3 as Figure 33 shows, fuel consumption with • AVL 489 to measure PN “Standard” profile is significantly better than fuel • AVL Indimodul for in-cylinder pressure (6 consumption with “Early Miller” profile (“Hard Miller” channels) in the figure), with an improvement of around 4% all • Eurins AdaMo control system to manage the test over the full load curve (area A). It was therefore bench. decided to continue testing using the “Standard” profile only. Following FPT’s best practice, prior to the proto assembly, the cylinder head was machined to install Figure 33 also shows the fuel consumption curve for one pressure sensor in each cylinder, to monitor the the “Standard” profile with the addition of EGR: in this combustion process, and a total of 8 thermocouples, case, fuel consumption is further reduced, as to keep the thermal condition of the head under predicted by 1D simulations and confirmed by SCE control: moreover, upon the installation on test bench, testing (Area B). On the overall, the best configuration the whole engine was instrumented in order to (“Standard” profile with EGR) gives an advantage in completely monitor the various parameters during the fuel consumption ranging between 5 and 8% on the testing. full load curve if compared to “Early Miller” profile without EGR. Engine testing The first operation performed on the engine was the debug of the new components and systems as well as the control system, both HW and SW and, once this operation was successfully completed, the engine was run-in.

Following this, it was possible to test the maximum performance of the engine. The goal of the project was to improve by 10% power and torque compared to the 2013 state-of-the-art NG engines which translated in 2200 Nm and 370 kW, target that was reached, as it is shown in Figure 32.

2800 Actual Figure 33 Fuel consumption in different engine 2600 Target configurations 2400 2200 The optimization process of the engine in steady state 2000 conditions followed: this process is aimed at finding

1800 the combination of the various factors that affect the engine's performance (ignition timing, injection Torque [Nm] Torque 1600 phasing, air control through throttle valve and WG, 1400 EGR valve opening) in order to identify the conditions 1200 that enable to achieve the optimum in terms of fuel

1000 consumption, combustion stability, margin vs. knock and emissions. 800 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000 Engine speed [rpm] The analysis was conducted over the same four operative points tested by AVL on the SCE: Figure 32 HDGAS performance, target vs. actual • Operating Point 1 (1200 rpm-8 bar BMEP) • Operating Point 2 (1000 rpm-2200 Nm)

52 ICPC 2019 – 2.3

• Operating Point 3 (1200 rpm-2200 Nm) • Operating Point 4 (1900 rpm-370 kW)

The experiment was designed to check, at first, the effect of each single innovative technology, then the combined effects of the various technologies employed.

First of all, EOI sweep was performed. The use of Direct Injection involves restrictions on the injection phase, namely:

• Injection must start after EVC, to avoid fuel short circuit to the exhaust • Injection may not be delayed indefinitely, to Figure 35 Spark timing effect on BSFC, Operating prevent subsonic flow through the injector. Point 1

The consequence of these constraints is a limitation Engine optimization led to excellent BSFC values in in the maximum duration of injection. the whole map, with efficiency exceeding 40% in various operating conditions. Retarding EOI leads to a faster combustion, as reported in which shows how MBF50 is anticipated At the end of the project, the engine was capable of with delayed EOI. performing a complete WHTC, in order demonstrate a 10% CO2 reduction on hot WHTC vs. 2013 state-of- the-art engine: the engine reached a 12% reduction in CO2 emission, therefore exceeding the target. Compared to a more up-to-date gas engine as the Cursor13 NG, the HDGAS engine showed a 4% improvement in fuel consumption over the same WHTC.

CONCLUSIONS

Natural gas engines can help reach the target for CO2 reduction, especially when used with bio-methane; moreover, they are readily available and their diffusion is growing, mainly in long haul missions.

In order to improve the thermal efficiency of gas engines (and, thus, further reduce GHG emissions), Figure 34 MBF 50 (average over 6 cylinders) vs. FPT joined the HDGAS project, to design and test an EOI innovative engine, specifically conceived to work witn natural gas, and incorporating several advanced Then, with the optimum EOI, spark advance sweep features, many of which used for the first time on an was carried out: starting from very retarded spark HD engine. timing (IMEP covariance around 6), spark timing was advanced to reach the Knock Limiting Spark Advance Working in close cooperation with other Partners (KLSA). Finally, sweeps of the Corona system (AVL, Politecnico di Milano and TUG), the task was parameters (first ignition voltage and then duration) accomplished employing advanced state-of-the-art was performed. Figure 35 shows the effect of the 1D and CFD calculations to complement the design, spark sweep on fuel consumption at Operating Point and SCE testing to confirm calculation results: the 1 with minimum BSFC reached at 8 deg CA after outcome is an engine which reached the project’s TDC. targets, showing a significant reduction in fuel consumption compared to the best natural gas engine currently available.

53 ICPC 2019 – 2.3

ACKNOWLEDGMENTS [10] http://www.ansa.it/canale_motori/notizie/eco_mo bilita/2018/10/25/iveco-stralis-a-metano-da- The Authors would like to thank the other Partners of record-1.728-km-con-un-pieno_1217b6ee-8fcd- WP4 of the HDGAS project for their support. 4087-8804-ac31f3f3cfd2.html [11] Krähenbühl P. et al., “FPT Industrial’s Leadership The research leading to these results received in Natural Gas Technologies for Industrial funding from the European Community’s Horizon Engines”, 38th Internationales Wiener 2020 Program under grant agreement 653391 Motorensymposium, Wien, 2017 (HDGAS PROJECT). [12] Heywood J. B., Internal Combustion Engine Fundamentals, McGraw Hill Inc., New York, 1988 [13] Fasching P., Natural Gas as Fuel for Monovalent and Dual Fuel Combustion Engines - an Experimental and Numerical Study, Ph.D. thesis, Technische Universität Graz, Graz, 2017 REFERENCES [14] Figer, G et al., "Nutzfahrzeug-Gasmotoren mit Dieseleffizienz", MTZ – Motortechnische [1] Ellingsen, L. A-W., Hung, C. R., Research for Zeitschrift, 75, (10), 2014, doi:10.1007/s35146- TRAN Committee – Resources, energy, and 014-0573-4 lifecycle greenhouse gas emission aspects of [15] Geiger, J. et al., "Der Erdgasmotor als electric vehicles, European Parliament, Policy Nutzfahrzeugantrieb – Trends und Department for Structural and Cohesion Policies, Herausforderungen bei der Entwicklung", 8th Brussels, 2018, doi: 10.2861/944056 – Internet Conference on Gas-Powered Vehicles, Stuttgart, http://bit.ly/2HDKk0y 2013 [2] IEA (International Energy Agency), The Future of [16] Lucchini T. et al., "Multi-dimensional modelling of Trucks: Implications for energy and the the air/fuel mixture formation process in a PFI environment, 2017 engine for motorcycle applications," SAE [3] Möhring L., Andersen J.: “CNG Mobility – Technical Paper 2009-24-0015, 2009, doi: Scalable, Affordable and Readily Available 10.4271/2009-240015 Solution for Environmental and Climate [17] Montanaro A. et al., "Experimental Challenges”, 38th Internationales Wiener Characterization of High-Pressure Impinging Motorensymposium, Wien, 2017 Sprays for CFD Modelling of GDI Engines," SAE [4] RICARDO, Impact Analysis of Mass EV Adoption Int. J. Engines 4 (1):747-763, 2011, doi: and Low Carbon Intensity Fuels Scenarios, report 10.4271/2011-01-0685 on behalf of CONCAWE, 2108 [18] Lucchini T. et al., "Automatic Mesh Generation for [5] https://ec.europa.eu/eurostat/statistics- CFD Simulations of Direct-Injection Engines," explained/index.php/Electricity_generation_stati SAE Technical Paper 2015-01-0376, 2015, doi: stics_%E2%80%93_first_results 10.4271/2015-01-0376 [6] CARB (California Environmental Protection [19] Paredi D. et al., "Gas Exchange and Injection Agency Air Resource Board), Joint agency staff Modelling of an Advanced Natural Gas Engine for report on Assembly Bill 8: 2016 assessment of Heavy Duty Applications," SAE Technical Paper time and cost needed to attain 100 hydrogen 2017-24-0026, 2017, doi: 10.4271/2017-24-0026 refueling stations in California, 2017 – Internet [20] Scarcelli R. et al., “CFD and optical investigations https://www.energy.ca.gov/2017publications/CE of fluid dynamics and mixture formation in a DI- C-600-2017-002/CEC-600-2017-002.pdf H2 ICE” ASME, ICEF2010-35084, 2010 [7] https://www.acea.be/press-releases/article/truck- [21] Lucchini T. et al., "Full-Cycle CFD Modelling of co2-targets-no-public-charging-points-for- Air/Fuel Mixing Process in an Optically electric-or-hydrogen-trucks Accessible GDI Engine," SAE Int. J. Engines [8] Semin R., “A Technical Review of Compressed 6(3):1610-1625, 2013, doi: 10.4271/2013-24- Natural Gas as an Alternative Fuel for Internal 0024 Combustion Engines”, American J. of [22] Golini S. et al., “Natural Gas Engines for Long- Engineering and Applied Sciences 1 (4): 302- Haulage Applications: Current Approach and 311, 2008 Future Developments”, 16th Conference: The [9] Schuller O. et al., Greenhouse Gas Intensity of Working Process of the Internal Combustion Natural Gas, Thinkstep report on behalf of NGVA Engine, Graz, 2017 Europe, 2017

54 ICPC 2019 – 2.3

DEFINITIONS, ACRONYMS, ABBREVIATIONS

AFR Air/Fuel Ratio ATS After-Treatment System BDC Bottom Dead Centre BEV Battery Electtric Vehicle BMEP Brake Mean Effective Pressure CA Crank Angle CFD Computational Fluid Dynamics CNG Compressed Natural Gas DI Direct Injection EGR Exhaust Gas Recirculation EOI End Of Injection EU European Union FEA Finite Element Analysis FC Fuel Cells FCV Fuel Cells Vehicle GHG Greenhouse Gas HC Unburned Hydrocarbons HD Heavy Duty HDV Heavy Duty Vehicle HI Homogeneity Index HVA Hydraulic Valve-lash Adjustement ICE Internal Combustion Engine IMEP Indicated Mean Effective Pressure IVC Intake Valve Closing KLSA Knock Limiting Spark Advance LCV Light Commercial Vehicles LH Long Haul LNG Liquefied Natural Gas NG Natural Gas MFB Mass Fraction Burned MCE Multi-Cylinder Engine SCE Single-Cylinder Engine SOI Start Of Injection TDC Top Dead Centre TKE Turbulent Kinetic Energy TR Tumble Ratio WG Wate Gate WHTC World Harmonised Transient Cycle

55

ICPC 2019 – 2.4 Potentials for friction reduction with commercial vehicle engines – Contribution of the power cell unit

Dr.-Ing. Andreas Pfeifer Mahle GmbH

M.Sc.Tobias Funk, Dr.-Ing. Thomas Deuß, Dipl.-Ing. Holger Ehnis Mahle International GmbH

Copyright © 2019 AVL List GmbH, Mahle GmbH and SAE International

ABSTRACT expected reduction of 20%, relative to emissions from 2019, is planned by 2025. A further 15% reduction is The reduction of friction losses in combustion engines scheduled by 2030. To achieve these objectives, the is of great significance for lowering fuel consumption reduction of friction inside the engine is highly and CO2 emissions. Extensive parameter studies of significant. With the goal of determining friction losses diesel and gasoline engines for passenger cars have under real conditions, that is, fired operating demonstrated that the piston group has significant conditions, MAHLE set up a friction power test bench potential for reducing friction. There are currently no for passenger car engines over ten years ago. In statutory limits on CO2 emissions for commercial extensive individual parameter studies on passenger vehicle engines, but in the future, however, there will car diesel engines and gasoline engines, with a total be strict limits in place for these types of engines as of over one hundred test variants, significant potential well. Friction reduction also plays a critical role for the for improvement was shown. end customer in the commercial vehicle sector, as fuel consumption makes up a significant proportion of Another friction power test bench has now been set the total cost of ownership (TCO). For these reasons, up for commercial vehicle engines so that potential an additional friction power test bench has been set savings can be determined by means of a systematic up at MAHLE to perform measurements on fired parameter study for the commercial vehicle sector as commercial vehicle engines. well.

This article presents the test bench setup, MEASUREMENT METHOD FOR measurement and evaluation methods, and the DETERMINING FRICTION LOSSES IN established measurement accuracy. The influence of THE COMMERCIAL VEHICLE ENGINE piston installation clearance, piston profile, and piston skirt surface structure on friction losses are assessed. Test bench setup Additionally, an optimized variant consisting of a combination of several individual parameters is Friction losses are calculated using the indication investigated to evaluate the overall potential for method. The indicated mean effective pressure IMEP reducing friction. To determine the savings in fuel is determined by means of high-pressure indication. consumption and CO2 emissions for each variant, a The brake mean effective pressure BMEP can be driving cycle simulation is performed. The resulting calculated from a torque measurement. The friction savings in CO2 emissions are up to 5 gCO2/km. mean effective pressure FMEP being sought is the difference between the indicated and brake mean INTRODUCTION effective pressure. In order to achieve the high level of measurement accuracy required for the indication method, external conditioning systems are used both Statutory limits for the CO2 emissions of passenger cars have been in place for a long time now and are for the coolant and for the engine oil. The friction continuously being made stricter. Statutory limits on power test bench is also equipped with conditioning systems for ambient air, intake air and fuel. A detailed CO2 emissions are now going to be introduced for heavy-duty commercial vehicles in the EU as well. An description of the test setup and measurement equipment can be found in [1].

56 ICPC 2019 – 2.4

Measurement and evaluation method three successive days are used. The average from the evaluation of the standard deviations results in To evaluate the saving potentials of individual reproducibility of ±0.01 bar for the FMEP. parameters for the piston group, friction measurements are performed under steady-state It is typically necessary to assemble an engine, to be conditions across the entire operating map. Design of able to compare different design variants. To evaluate experiments (DoE) is used to determine the FMEP measurement accuracy, the influence of engine maps. Using 21 operating points that vary the assembly on FMEP must therefore be investigated. parameters of load, engine speed, and engine To this end, the engine is fully disassembled and temperature, a FMEP map is established for the reassembled several times using the identical entire operating range of the engine. Three such components and friction power measurements are operating map measurements are performed for each taken in between each disassembly and reassembly. variant. The FMEP map for a variant is the average The average standard deviation for FMEP is ±0.02 of these three measurements (Figure 1, left). The bar, which is designated the confidence limit. influence of such a design variant on engine friction is depicted in the form of a FMEP difference map The FMEP differences must be outside of the (Figure 1, right). To obtain this, one of the FMEP confidence limit in order to unambiguously attribute maps for the variants to be compared is subtracted the result to the design change. If the FMEP from the other. This difference map can be used to differences are below the confidence limit, then draw conclusions about friction behavior across the additional evaluations must be performed in order to entire operating map. be able to draw conclusions at least about tendencies. A detailed description of these tests can be found in [1].

Figure 1 Generation of difference map: FMEP maps of two variants to be compared (left), and the resulting FMEP difference map (right)

Measurement accuracy Figure 2 Measurement accuracy of FMEP measurements: repeatability, For the indication method, very high measurement reproducibility, and influence of engine accuracy is critical to resolve the very small assembly differences between the indicated and brake mean effective pressure. Therefore, extensive investigations of measurement accuracy are RESULTS OBTAINED FROM THE performed prior to the actual parameter study. They COMMERCIAL VEHICLE ENGINE cover repeatability, reproducibility, and the influence of engine assembly on measurement accuracy. In a systematic parameter study, the influence of individual piston group design parameters on friction The results for the 21 operating points from the FMEP was investigated. Table 1 shows all of the parameters map are used to evaluate individual measurement that have been varied for the commercial vehicle accuracies. To evaluate repeatability, the standard engine. In over 30 tests, the influence of different deviation of the FMEP is calculated from ten design parameters including the tangential force of successive individual measurements. The average the oil control ring, the piston type, and the geometry results in repeatability of ±0.002 bar for the FMEP of the connecting rod small end have been examined (Figure 2). [2]. The results for the parameters of piston installation clearance, piston profile, and piston skirt The reproducibility is also evaluated on the basis of surface structure are presented below. the standard deviation for the FMEP. In contrast to repeatability, operating map measurements from

57 ICPC 2019 – 2.4

Piston group component Parameter Installation clearance Pin offset Piston profile Skirt roughness Piston Skirt coating Skirt area Pin bore geometry Piston type Tangential force of oil control ring Rings Top ring height Pin Pin coating Other Oil viscosity

Table 1: Parameters investigated for commercial vehicle engine Piston installation clearance Investigations on the passenger car diesel engine with aluminum pistons have shown that piston installation clearance can have significant potential for reducing friction [3]. Based on these findings, four different installation clearances are investigated for the commercial vehicle engine with steel pistons. The baseline variant has an installation clearance of 63 µm, measured on the skirt coating after the test run. One variant with installation clearance reduced to 31 µm and two variants with increased clearances of 85 µm and 107 µm are tested. The corresponding FMEP difference maps are shown in Figure 3. It is evident that variation of installation clearance leads to only minor differences in FMEP. No distinct dependence on load or engine speed is observed. The results are very close to the confidence limit, but nevertheless the three difference maps show that reducing installation clearance tends to cause greater friction with steel pistons as well.

The reason for this fundamentally small influence of the installation clearance is presumably the very similar material properties of the cylinder liner and the piston. In contrast to many passenger car diesel engines with aluminum pistons, the commercial vehicle engine has steel pistons that are installed in so-called wet cylinder liners. The piston and liner therefore have similar thermal expansion coefficients. This means that even under full load conditions, there can presumably be no interference between the piston skirt and the cylinder wall. Even pistons with small installation clearance still have sufficient operating clearance in the warm state and thus have only a slight disadvantage in terms of friction losses.

Figure 3 FMEP difference maps, measured during operation with various piston installation clearances (top: 63 µm – 31 µm; center: 63 µm – 85 µm; bottom: 63 µm – 107 µm), engine temperature 100 °C

58 ICPC 2019 – 2.4

Piston profile The piston profile can affect the build-up of the lubricating film on the piston skirt. Starting with a conventional, barrel-shaped, oval piston profile, three more different profiles are investigated. These alternative piston profiles are designed with the goal of influencing lubricating oil distribution on the piston skirt such that hydrodynamic lubrication conditions prevail to a greater extent. The tested piston profiles are shown in schematic form in Figure 4.

Figure 5 indicates the corresponding FMEP difference maps. The Camel Back 1 piston profile, with pronounced horizontal humps, shows significant benefits in FMEP over the entire operating map (Figure 5, top). No dependence on engine speed is evident, although there is a clear dependence on load. The greatest advantages in FMEP are evident at higher loads across the entire speed range. The maximum savings in FMEP are 0.09 bar.

The Camel Back 2 piston profile, with more substantial humps, shows similar friction behavior to the Camel Back 1 (Figure 5, center). No significant dependence on engine speed is evident. The evident dependence on load, however, is less pronounced than for the Camel Back 1 profile. The potential savings in FMEP are also somewhat lower at about 0.05 bar.

The friction behavior of the Banana Shape piston profile, with the humps curved like a banana, is largely the same as that of the Camel Back 2 (Figure 5, bottom). No benefit in FMEP is evident in the low to medium load range. The maximum FMEP benefit is Figure 4 Schematic description of the examined also 0.05 bar. piston profiles

The mixed lubrication component of piston skirt friction increases as the load increases, because greater lateral forces arise under these operating conditions and the thermal expansion of the components increases, especially that of the piston. The thermal expansion increases the contact pressure between the piston and cylinder wall. When the piston profile improves the lubricating oil distribution on the piston skirt, this promotes hydrodynamic lubrication conditions and the mixed lubrication component decreases. For all of the investigated piston profiles, therefore, the greatest friction advantage is seen at high loads. The Camel Back 1 piston profile is the optimization with the greatest friction advantage.

59 ICPC 2019 – 2.4

Piston skirt surface structure To investigate the influence of surface structure on friction, two different piston skirt surface profiles are compared. Sketches of the two surfaces are shown in Figure 6. The base variant is the conventional series production design of the skirt surface. The modified surface structure is characterized by distinct longitudinal waviness. Thus small plateaus form after test run whereby the effective running surface is reduced in comparison to the conventional piston skirt surface structure.

Figure 6 Schematic description of the examined piston skirt surface structures after test run

The FMEP difference map in Figure 7 shows the influence of the modified piston skirt surface structure on FMEP. Small friction advantages are evident in the low to medium load range. As the load increases, however, friction disadvantages appear for the modified, rough surface structure. These disadvantages are especially pronounced at low engine speeds and high loads. This operating range is characterized both by increased component temperatures and therefore reduced operating clearance and by high lateral forces. This results in high surface pressures for the modified surface structure. The mixed lubrication component consequently increases.

Figure 5 FMEP difference maps, measured during operation with different piston profiles (top: Camel Back 1; center: Camel Back 2; bottom: Banana Shape), engine temperature 100 °C

60 ICPC 2019 – 2.4

400 0.7 350 0.6 300

/year] 0.5 € 250 0.4 200 0.3 150 100 0.2

TCO savings [ savings TCO 50 0.1

0 0.0 Fuel Consumption Savings [%] Savings ConsumptionFuel

Figure 8 Fuel consumption savings and TCO savings from the parameter Camel Back 1 piston profile Figure 7 FMEP difference map, measured during The average values for potential savings in CO2 operation with different piston skirt emissions for the investigated parameters are shown surface profiles (conventional - modified in Figure 9. The parameter piston installation piston skirt surface structure), engine clearance has potential savings of up to 2.5 gCO2/km. temperature 100 °C By optimizing the piston profile, CO2 emissions can be reduced by up to 3 gCO2/km, while a suitable POTENTIAL FOR CO2 REDUCTION selection of the surface profile on the piston skirt has a maximum potential of 1.3 gCO2/km. The findings of In addition to the potential savings in FMEP, great the previous individual parameter studies are interest lies in the determination of the associated fuel incorporated into a so-called best of variant. The consumption savings and reduction in CO2 emissions. combination of various design parameters enables The measured FMEP maps, together with the CO2 emissions savings of about 5 gCO2/km. corresponding vehicle data, are used as input parameters for a driving cycle simulation. Three 5.0 different driving cycles (Long Haul Vecto, HHDDT 4.0 transient, HHDDT cruise) are considered, with three 3.0 different vehicle load cases for each (empty, reference, full). This results in nine individual values 2.0 for evaluating friction-based fuel consumption 1.0 savings and reductions in CO2 emissions. The 0.0 influence of the dynamic friction behavior on the -1.0

cumulative fuel consumption and CO2 emissions in

Emission Saving [g/km] Saving Emission 2

-2.0 Modified

- 31 µm 31

driving cycles is considered in [4]. µm 85

107 µm 107 µm 107

CamelBack 1 CamelBack 2

-

Banana Shape

-

CO

- -

- -

-3.0 -

Best Best of

63 µm 63

Piston Piston Profile Piston Profile Piston Profile

63 µm 63 63 µm 63

Fuel costs represent a large portion of the total cost -4.0 µm 31 Conventional

of ownership (TCO) in the commercial vehicle sector. Piston Design Parameters

Conventional Conventional

Conventional

Piston Installation Installation Piston Clearance Piston Installation Piston Installation Clearance Piston Installation Clearance Piston Installation Clearance

-5.0 Piston Skirt Surface Structure Figure 8 shows the percentage of fuel savings and associated potential annual cost savings for the Figure 9 Fuel consumption savings and CO2 example of the Camel Back 1 piston profile results. emissions reduction due to reduced The cost estimate is based on annual mileage of friction (values are averages from three 150,000 km and average fuel consumption of driving cycles (Long Haul VECTO, 34 L/100 km. A price for diesel of EUR 1.20 per liter HHDDT transient, HHDDT cruise) with was assumed. The resulting cost savings are three load cases for each (empty, between EUR 100 and EUR 350 per year, depending reference, full)) on the driving cycle and load state considered.

61 ICPC 2019 – 2.4

CONCLUSION

The measurement and evaluation procedure that has been used for friction power measurements on passenger car engines was successfully transferred to the commercial vehicle engine. A high measurement accuracy of ±0.02 bar FMEP is possible for the engine used. As part of a systematic individual parameter study, the parameters piston installation clearance, piston profile, and piston skirt surface structure have already been investigated. Further tests will determine the influence on friction of the parameters structural stiffness, ovality, skirt surface, and coatings. With the findings to date, CO2 emissions can be reduced by up to 5 g/km.

REFERENCES

[1] Deuß, T.; Ehnis, H.; Schulze Temming, R.; Künzel, R.: Friction power measurements with a fired HDD engine—method and initial results. 17th Stuttgart International Symposium on Automotive and Engine Technology, 2017. [2] Deuß, T.; Ehnis, H.; Schulze Temming, R.; Künzel, R.: Friction power measurements with a fired commercial vehicle engine—piston group potentials. MTZ Motortechnische Zeitschrift. 02/2019, Volume 80. [3] Deuß, T.; Ehnis, H.; Freier, R.; Künzel, R.: Friction power measurements of a fired diesel engine—piston group potentials. MTZ Motortechnische Zeitschrift. 05/2010, Volume 71. [4] Funk, T.; Ehnis, H.; Künzel, R.; Bargende, M.: Dynamic friction behavior of a gasoline engine in transient operation. 19th Stuttgart International Symposium on Automotive and Engine Technology, 2019.

62

ICPC 2019 – 2.5 ICE optimized for off-road hybrid powertrain

Dr. - Ing. Markus Schwaderlapp, Dr.-Ing. Paul Grzeschik DEUTZ AG

Copyright © 2019 AVL List GmbH, DEUTZ AG and SAE International

ABSTRACT see Figure 2: regenerative fuels and electrification. DEUTZ is very active in following both roadmaps.

This paper covers the DEUTZ strategy towards CO2 free powertrains. The discussion is focused on hybrid powertrains and on an optimized engine layout for these applications: reduction of complexity can result in downsizing by reducing the cylinder number or adjusting the technological level of the engine. The additional low-end torque provided by the electric motor also provides the opportunity to replace a Diesel engine by a cost-effective gas engine. In this context, the goal is always to make use of the already existing DEUTZ portfolio to maximize the commercial competitiveness of hybrid applications.

INTRODUCTION Figure 2 Approaches towards a CO2 neutral The off-road powertrain is a domain of the internal drivetrain combustion engine, see Figure 1: it combines in an almost ideal manner compactness, high operation Electrification means full electric drives for smaller range and robustness. With Stage V the pollutant vehicles and hybrid solutions for higher demands of emissions are reduced to a minimum contributing to power and operating range. Combining an engine a clean environment. with a 48 V electric motor (“mild hybrid”) or a 360 V motor (“full hybrid”) saves fuel (CO2) and has the advantage of an improved transient response. For the combustion engine, hybrid applications create the potential for component simplification and reduction of engine complexity.

In the end, the technical solution must create a benefit for the end customer – either by reducing the total cost of ownership or by creating advantages for the application of the vehicle, such as local zero emission or low noise in electrical drive mode. The hybrid solutions currently under development at DEUTZ offer those new potentials for the customer.

HYBRID POWERTRAINS FOR NRMM Figure 1 Application examples for DEUTZ engines One of the characteristics of NRMM powertrains is However, even with highly efficient Diesel engines, the wide variety of applications and operation profiles, CO2 minimization remains a challenge. There are two see Figure 3. The technological and economic complementary roadmaps to convert an almost potential for electrification of these powertrains is emission free powertrain into a CO2 free powertrain, strongly dependent on factors like dynamics of

63 ICPC 2019 – 2.5 operation, power and energy demand, supply infrastructure availability, powertrain on/off ratio and others.

Figure 3 Typical engine operation profiles for different NRMM applications

Figure 4 Hybridization potential for different energy demand and operating range regimes

64 ICPC 2019 – 2.5

Looking to the different applications, Figure 4 shows background, it will not be economically and practically how the degree of electrification depends on the feasible to electrify these applications in the energy demand and the available energy supply foreseeable future, although several approaches to infrastructure, which is equivalent to the necessary do so exist in pre-develompment and prototype operating range of the respective application. status.

Machines with low cycle energy demand, which at In between these boundary cases, there is an same time do not have to travel too far away from re- intermediate area where a combination of electrical charging infrastructure, are most conveniently and conventional (i.e. ICE based) drivetrains is equipped with a fully electric powertrain. They do not becoming more and more attractive as technology require large, sophisticated and therefore high-cost levels – especially those of the energy storage device energy storage capabilities. For applications such as – progress. A large number of different applications indoor lift trucks, light to medium-duty fork lifts, or such as heavy-duty lift trucks, wheel loaders, airport towing tractors, a strong trend towards full telehandlers and, to some extent, excavators, fall into electrification has already set in in the past years, and this category. with the further optimization of electric powertrains, it can be expected that these applications will remain a Figure 5 shows the sub-systems of a typical 48 V domain of electric drives. hybridized powertrain for a telehandler application as developed by DEUTZ in the E-DEUTZ program. It is On the other end of the energy demand vs. operating clear that such a powertrain raises a number of range field, we have applications such as agricultural questions and challenges which are very new for both tractors which typically operate far from any high- manufacturers and operators of NRMM; such as the performance re-charging infrastructure. This is location, or distribution, of the energy storage, the combined with a very high demand for both power packaging space for electrical and electronical and energy to fulfill the task. For a day of e.g. equipment, the requirement for a low temperature ploughing duty, a tractor requires the energy cooling system, the modified weight distribution of the equivalent of approx. 600 L of Diesel fuel. To vehicle, personal safety considerations and so on. accommodate this amount of energy in a chemical Obviously, one of the most important and immediate storage device, a battery module with a volume of questions is that for the economical viablilty of a approx. 3500 L and a mass of around 10 t would be hybridized powertrain. necessary [1]. With the current technology

Figure 5 Typical layout of a hybridized 48 V NRMM powertrain

65 ICPC 2019 – 2.5

As for the shown powertrain, by careful total system inherent advantages of a hybrid drivetrain (better layout, the initial costs of electrification can be dynamics, zero pollutant and noise emission when partially or fully compensated. By downsizing of the driving fully electrically) are sufficient to convince combustion engine, the engine-out power drops owners and operators, the ICE optimized for below the regulatory border of 56 kW and allows for hybridization has to contribute to the cost- the subsequent elimination of the expensive SCR effectiveness of the whole hybrid powertrain. system. In addition, it can be shown that hybridization of a telehandler as performed by DEUTZ will yield an COMBUSTION ENGINES OPTIMIZED operating cost saving, depending on the usage cycle, FOR HYBRID POWERTRAINS of around 10,000 € during the lifetime of the machine. Nevertheless, the savings margin will not always be To achieve this goal, DEUTZ offers three main high enough to cover all technically reasonable approaches to the optimized combustion engine as applications. So apart from the cases where the showin in Figure 6.

PEAK LOADS COVERED BY EM: OCCASIONAL USE OF ICE: ICE AS RANGE EXTENDER: SIZE FIT GRADE FIT TECH FIT

Figure 6 Alternatives for hybridized ICE optimization approaches

The choice between the alternatives is mainly telehandler, a very convenient approach is so replace dependent on the operation scenario of the NRMM a four-cylinder engine (in this case, the DEUTZ TCD application and the hybrid powertrain operating 3.6 L 74 kW machine) by a cost-effective three- strategy. cylinder engine (DEUTZ TCD 2.2 L 56 kW) and bridge the power gap with an appropriately sized Size Fit electrical machine. As Figure 7 shows, this substitution additionally improves the total low-end Downsizing is a feasible approach where the bulk of torque of the drivetrain while at the same time shifting operation load and time share is still provided by the the application’s mean load curve into a higher-load, ICE while the electric drive is mainly used to boost for consumption-optimum range of the ICE operation peak performance events. In these cases, as map. demonstrated by the DEUTZ 48 V hybridized

66 ICPC 2019 – 2.5

Figure 7 Telehandler operation map for conventional and hybridized powertrains

Figure 8 Packaging comparison between conventional (TCD 3.6 four-cylinder engine, grey) and hybridized power plant (with TCD 2.2 three-cylinder engine, red)

An additional significant benefit of this specific reducing cost and installation space demand. The example is that legislation allows 56 kW engines to overlay in Figure 8 gives an impression of the go without sophisticated EAT system, thus further longitudinal size of the respective power plants. While

67 ICPC 2019 – 2.5 keeping the overall length, the hybridized powertrain effectivenesse. Figure 9 gives some examples of even leaves some formerly blocked space above the possible cost optimization potentials on sub-system clutch/gearbox assembly. and component level. This “right-grade” approach proves to be especially effective within the DEUTZ That space can be used to accommodate some of the strategy, which traditionally allows for competitive and necessary electrical equipment or, if the electrics can economically feasible versioning of components even be placed in a different location, allows for a size in small yearly quantities. For example, as showin in reduction of the engine bay – an important issue for Figure 10, the replacement of a steel crankshaft by a the telehandler vehicle design for reasons of visibility cast iron crankshaft could be amortized within a towards the right-hand side of the machine. yearly production rate of around 13,000 pieces. As soon as demand for these kinds of applications Grade Fit reaches a sufficiently high level, DEUTZ will engage in the described cost optimization. The typical NRMM combustion engine is an extremely robust heavy-duty machine designed for a long product life under demanding ambient Crankshaft Total Costs conditions. Depending on their displacement volumes, DEUTZ engines are certified for B10 life- times of 6,000 to 13,000 hours and even after their Invest Cast Iron “first life”, they can be refurbished for a “second life” as a cost-effective alternative in the DEUTZ XCHANGE program. Major components and systems Initial Steel are regularly found to remain in excellent condition even after long usage. However, in cases where the 0 5000 10000 15000 20000 electrical portion of a hybrid powertrain is designed to Pieces p.a. be the main power source (e.g. in 360 V full hybrid applications), the ICE does not have to be specified Figure 10 Break-even calculation for the for such long lifetime or heavy duty usage. introduction of a low-cost crankshaft

Turbocharger Technology Tech Fit Cylinder Head Material Hybridized applications which require even less Cam Tappets contribution by the ICE (e.g. those which use the ICE Piston Material just as a range extender) allow for application of cost- Piston Cooling System effective, ligher-duty engines such as gas engines, Conrod Bushing which are offered by DEUTZ in the form of the G2.2 Crank Shaft Material [2]. This engine, designed for LPG and, with an adaptation, to CNG is the first DEUTZ engine for Bearing Shells NRMM serving the market for material handling but Oil / Fuel Filters as well other applications demanding for a robust engine where a Diesel engine with turbocharger and a sophisticated emission control is not the primary option.

The engine, with its main technological features shown in Figure 11, offers a robust approach taking

into account economic features as well as a compact Figure 9 Cost optimization potential of a low design. With a high level of synergies to the diesel usage ICE in a hybridized powertrain engine platform of D/TD/TCD2.2, the installation in the same applications is redundant, providing a The powertrain’s operating strategy would employ the maximum torque of 160 Nm and a rated power of combustion engine only occasionally to either 42 kW at 2800 rpm. significantly increase overall power output or to cover occasional work scenarios when long-term electrical DEUTZ is also introducing the G2.9, a four-cylinder operation is not feasible. Consequently, individual variant of the same engine family.It is also planned to systems and components may be replaced with extend the gas engine business to higher lower-grade derivatives to improve the ICE’s cost- displacements to achieve ratings of 70 kW and more.

68 ICPC 2019 – 2.5

Pressure regulation valve (DEPR) • 2nd stage pressure controller Lock-Off Valve (electronic) Evaporator • Mounted at • Regulation of LPG/CNG mass flow • Converting liquid the gas tank • Adjusting air ratio 14.7:1 gas into gaseous form • Cuts the Gas supply after • Reducing system pressure ignition-off (high pressure) to low pressure • 1st stage pressure controller (mechanic) Intake air manifold Modified Cylinder head and Crankcase castings Throttle Modified Valves Piston (ε = 11): and Valve seats • Optimized piston bowl geometry • Valve pocket for inlet valve (Backfire) • Suction pressure at closed throttle down to 200mbar abs. LPG Mixer → oil consumption and reverse blow-by 3-way catalyst → U-flex design of oil ring allows adaptation to cylinder surface deformation • Friction reduced by U-flex ring, reduced ring thickness and reduced ring force

Figure 11 DEUTZ G2.2 technology overview

In this field, DEUTZ has been developing the This publicly funded pre-development project is HoLeGaMo high power gas engine with 150 kW and helping to extend the application range for gas engine 800 Nm [3], see Figure 12. – and with them that of cost-effective hybridized powertrains – into power ranges of intermediate to heavy machinery.

DEUTZ PORTFOLIO OUTLOOK

In the coming years, the E-DEUTZ approach will gradually address machines and applications with increasing power and energy demands as shown in Figure 13. Starting with ICE powers < 56 kW as seen in today’s telehandler demonstrators, on a 2 … 10 year scale, intermediate tractors and larger equipment such as excavators are in the focus of hybrid power trains developed and offered by DEUTZ.

In parallel, the development of full-electric drives for small to intermediate applications will be continued, expecting to cover full electric small construction and agricultural equipment > 56 kW before the year 2030. Figure 12 DEUTZ HoLeGaMo high power gas engine

69 ICPC 2019 – 2.5

Starters & selective replacement of Full replacement of mechanical Diesel downsizing for larger attachments attachments (e.g. tractor, etc.) equipment >160kW Diesel downsizing Diesel downsizing for larger for small equipment equipment >56kW (Roller, etc.) <56kW (fork lift truck, telehandlers, etc.)

Hybrid Today 2-10 year horizon

Full-Electric Today 2-10 year horizon

Compact equipment and material Small construction and Small construction and handling <37kW residential equipment <56kW agricultural equipment >56kW wheelloader, etc. Lawn, Mowers, etc. Telehandler, small tractors, excavators etc. Small rollers, Forklifts, etc. Compact utility tractor, etc.

Figure 13 Today’s and future applications for electrified powertrains

CONCLUSION REFERENCES

With the ever-increasing demand for both pollutant [1] [1] M. Schwaderlapp, M. Winkler, Th. emission reduction and CO2 minimization, powertrain Adermann, K.-P. Bark: CO2-neutrale Mobilität. electrification has become an important topic for Potenziale von alternativen Kraftstoffen und NRMM manufacturers and operators. As fully electric Elektrifizierung bei Off-Highway-Anwendungen. powertrains are not (yet) able to cover the full range In: MTZ 11/208 of applications, hybrid powertrains are becoming an [2] [2] H. Bülte, C. Funke, K.-P. Bark, K. Tedsen: important step towards CO2 neutral operation of non- DEUTZ G2.2 – The New 3-Cylinder Gas Engine road equipment and machinery. for Nonroad Mobile Machinery. In: Heavy-Duty-, On- and Off-Highway Engines. Future Hybridization, however, has to be accompanied by a Challenges. 13th International MTZ Conference strong effort to minimize cost of the conventional on Heavy-Duty Engines, Cologne 2018 components to make electrified powertrains as [3] [3] H. Bülte, G. Töpfer, C. Funke: Gas Engines economically attractive as possible. The “size fit” and for Mobile Machinery - A Contribution to the “tech fit” approaches are already part of the work Reduction of CO2 Emissions. In: 6th scope, while the “grade fit” approach will be International Engine Congress, Baden-Baden subsequently implemented when customer demands 2019 increase. With their strong, diversified engine portfolio, DEUTZ is confident to provide the optimum, cost-effective solution for the ICE in any application, DEFINITIONS, ACRONYMS, environment, and operating strategy. ABBREVIATIONS

ACKNOWLEDGMENTS HoLeGaMo High performance gas engine (Hochleistungs-Gasmotor) The HoLeGaMo project has been funded by the ICE Internal combustion engine German Federal Ministry for Economic Affairs and Energy. LT Low temperature NRMM Non-road mobile machinery PMSM Permanent-magnet synchronous motor

70

ICPC 2019 – 3.1 Fuel Cells: A Profitable Zero-Emission Solution for Heavy Duty Trucks

William Resende, Heimo Schreier, Dr. Alexander Schenk, Martin Ackerl AVL List GmbH

Copyright © 2019 AVL List GmbH and SAE International

ABSTRACT However, as this technology has been developed mainly for passenger car application in the past, some In this paper, a fuel cell powertrain for HD truck engineering challenges need to be solved to achieve application is evaluated in greater details. The market a sufficiently profitable and durable truck powertrain drivers are discussed, the technical challenges and hence, achieve a significantly high market described and solutions from AVL are proposed. penetration to contribute to the necessary emissions Specific focus is given to the adoption of the current reductions in the on-road transport sector. fuel cell technology to achieve a profitable zero- emission powertrain for heavy duty trucks. MARKET DRIVERS

INTRODUCTION In the trucking business, there are several factors driving the change to low or even zero emission The trucking industry is facing challenges with powertrains. Among them, government CO2 regards to powertrain emissions as governments regulations, drive bans and tolls. around the world push for stricter regulations, on a country and city level. These regulations are arising CO2 Emission Regulations from findings and recommendations from several Over the last years governments have been taking study groups which pin the temperature increases on strong instances with regards to greenhouse gas greenhouse gas emissions. The Paris Agreement emissions (see Japan, China, Korea, European was the culmination of those discussions and set Union and California). Figure 1 summarizes the clear targets for CO2 reductions across different current CO2 regulations and fuel economy standards sectors. To achieve the upcoming CO2 targets, in place in several countries for commercial vehicles. vehicle OEMs need to commercialize alternative powertrain technologies like hybrids as well as battery and fuel cell electric powertrains. Specifically the long-haul application is in the focus, as this vehicle category contributes most to the emissions of the on- road transport sector.

Additionally, city governments have started to propose bans to combustion engine powertrains from urban areas to avoid local air pollution and noise emissions. To cope with this demand battery electric powertrains have been brought to the market for trucks as well as busses. However, one of the Figure 1 CO2 and Fuel Economy Regulations in remaining challenges of this technology is the limited the World today range as well as the typically longer refilling times. These regulations are expected to get stricter over A technology which has the potential to overcome the next years as the participation of CO2 emissions these challenges is fuel cell technology. During the from the trucking business will start representing a past years this technology has made significant larger share of the total amount of emissions. This is progress and different OEMs have already been shown in Figure 2. developing and testing it for commercial vehicles.

71 ICPC 2019 – 3.1

and trucks from road tolls. Based on current toll rates, a 40-ton truck (Euro VI) toll costs 18.7 EUR per 100 km. Considering 120,000 to 240,000 km driven annually on German roads, 22,440 to 44,880 EUR of operating expenses would be saved per truck and year.

In Switzerland, the Swiss federal customs charge commercial vehicle based on the total weight, emission level and kilometers driven in Switzerland and the principality of Liechtenstein (HVC federal tax, i.e. the performance-related heavy vehicle charge). Trucks with electric powertrains are exempt from this HVC federal tax. For example, the road toll charge for a 40-ton diesel truck (Euro VI) would be 91.20 CHF Figure 2 Status and Forecast of Green House Gas (81.37 EUR) per 100 km. Considering an annual Emissions contribution for different mileage of between 120,000 to 240,000 km, the toll transportation segments (Source: costs would be approximately 109,000 CHF and Transport Environment, EEA, EC, Roland 219,000 CHF respectively (approx. 97,000 € and Berger) 194,000 €).

Due to this, at the end of 2018, the European FUEL CELL TECHNOLOGY parliament has approved a corresponding law that determines an interim target of 30% reduction of CO2 The trucking business rests on the profit potential that emissions by 2030 (in comparison to 2019). A can be made by the cost of operating a truck and the mandatory target of 15% by 2025 (in comparison to corresponding amount received to transport the 2019) was also agreed upon. goods. The TCO of a truck is composed of several factors, for example depreciation, fuel costs, Internal Combustion Engine Bans insurance, maintenance, opportunity cost, taxes, tolls, etc. To maximize profit with a truck, its revenues, Cities and provinces have started a strong campaign uptime and freight utilization need to be maximized to ban internal combustion engines as they consider while minimizing operating costs. These parameters them to be the major source of particulates causing provide the answer why fuel cells can offer a respiratory and other diseases. As of February 2019, profitable alternative for the trucking business while the number of cities and regions globally planning a being zero emission. The reasons why fuel cells are ban amounted to 30. a good alternative, are explained in the next chapters. Even though some of these bans are still not Refilling Time approved in the legislative and the courts, it shows the direction that several cities are willing to start Based on the current refueling infrastructure and imposing these types of bans. refilling times, the amount of km per min that can be refilled for different fuels is compiled in Table 1. Additionally, several cities have started to restrict night deliveries into their cities with focus on noise Fuel type Refueling Comments emissions. Therefore, the need for zero emission and speed zero noise truck powertrains is imperative to solve Diesel 400 km/min 130 L/min* this new paradigm. 32 L/100 km** Electricity (250 kW) 1.9 km/min 250 kW* 220 kWh/100 km ** Tolls Electricity (1 MW) 7.6 km/min 1 MW* 220 kWh/100 km** In October 2018, the European parliament backed a Hydrogen 10 km/min 1 kg of H2/min* 50% discount on road toll charges for zero emission 700 bar 10 kg of H2/100 km** trucks. The new regulation needs approval from the today Hydrogen 700 bar 100 km/min 10 kg of H2/min* member states before taking effect. Once the (H70HF Standard) 10 kg of H2/100 km** regulation is enforced, the toll, which currently depends on the tonnage, fuel and emission type will * Dispensing rate or power level see another penalty on internal combustion engines. ** Fuel Consumption or Energy Consumption

Also, in October, the German government Table 1 Refiling Time comparison for Diesel, Fuel communicated that they will exempt electric buses Cell and Batteries

72 ICPC 2019 – 3.1

The underlying assumptions (i.e. typical fuel refilling Using these numbers in a TCO model, AVL rates and vehicle consumption) for the calculation are determined the TCO of the different truck noted in the last column. powertrains. The results are shown in Figure 3.

As one can see, diesel refilling rates are faster than those that can be achieved with refilling/re-charging technologies for fuel cell/battery trucks today. Incumbent diesel technology is about 40 times faster than today’s hydrogen refilling technology, which is +38%

about 1 kg of hydrogen per min due to regulatory TCO standard limitations and tank technology (which +7% dictates the size of nozzles and dispensing rate). Recently a consortium of companies was formed and plans to change the standards so that higher refueling rates can be achieved, as shown in the table above. When this body of work is completed, fuel cell trucks Diesel Fuel Cell Battery could be refilled with up to 100 km/min, requiring only 10 min to be filled up for a 1,000 km trip. Battery Figure 3 TCO comparison of diesel, fuel cell and charging capabilities are also evolving. Some battery trucks companies are now evaluating trucks to be charged with a power of 1 MW, which would be enough to reach 500 km of range in 1 hr. The graph in Figure 3 indicates that fuel cell truck TCO would likely be slightly higher than diesel trucks Assuming the best refilling rates for fuel cell and (7%) but could be cheaper in case hydrogen fuel battery, these trucks would be able to have an annual reaches the best case scenario costs (4 €/kg). Battery mileage similar to diesel for most use cases. This trucks remain more expensive than diesel and fuel cell trucks if one were to take the average scenario. assumption was made for a typical use case where a Battery trucks could have a similar TCO to fuel cells long-haul HD truck drives on average 800 km per day in case hydrogen costs do not fall below 8 €/kg and with about 1 hour stop for refueling and others. This energy prices at the charger are equal or less than assumption will vary depending on use cases, driver 0.30 €/kWh. The cost breakdown can be seen in availability and country regulations. Figure 4 below (Mid Scenario): Total cost of Ownership (TCO) 5% 6% 4% 3% 1% 3% 1% 4% The TCO of a truck can be calculated using the costs 1% 0% 0% figures as summarized in Table 2 22 18% 35% 29 % % 55% 60% 10% of total purchasing costs per 33 % Depreciation Costs year for 10 years 5% 8% 6% Diesel Fuel Cell Battery Electricity: 0.30 to 0.50 €/kWh Fuel Costs H2 Cost: 4.0 to 8.0 €/kg Diesel: 0.90 to 1.20 €/L

Insurance Costs 2,660 € for all powertrains Figure 4 TCO break down for different powertrains

Diesel and Fuel Cell: 10% of As one can notice, fuel costs are dominant for the 3 powertrain purchasing costs Maintenance Costs powertrains, especially for fuel cells and batteries. Battery: 5% of powertrain purchasing costs Second biggest cost driver is the driver. Diesel has an additional toll cost, which is the third biggest cost 3% Interest per year applied to Opportunity Cost driver. Powertrain costs, maintenance and insurance total purchasing costs play a secondary role in the TCO. 18 €/100 Km for Diesel Trucks Toll Costs Zero Emission Trucks exempt Payload Driver 60,000 €/year Another important measure for the trucking business is the total payload that a truck can transport. Trucks Table 2 Assumptions for TCO Calculation are usually rated in terms of total allowed tonnage that they can have while transiting on public highways. Therefore, it is in the interest of the business that the

73 ICPC 2019 – 3.1 weight of truck and powertrain should be minimized will discuss the most critical of them: cost, durability, in order to maximize the payload that can be packaging and cooling. transported. Figure 5 shows a weight comparison of the different powertrains (with current technology and Cost target technology). A fuel cell powertrain is made up of several components and their costs are shown below in 10.000 Figure 6 in comparison with diesel and battery 9.000 powertrains. The assumptions for the cost 8.000 7.000 calculations are shown in Table 3 6.000 5.000 Diesel 500 HP Engine 4.000

3.000 Weight (kg) Weight 2.000 1,000 km Range 1.000 0 Fuel Cell 300 kW Fuel Cell (3 X 100 kW Diesel Fuel Cell Battery Diesel Fuel Cell Battery Modules) @ 40 €/kW Today Future Potential 500 km Range (52 kg of H2) FCS Tank + Fuel Battery Engine Cooling E/E Rest 50 kWh Battery @ 100 €/kWh Figure 5 Powertrain weight comparison Battery 500 km Range

As it can be seen, fuel cell trucks can already show 1,100 kWh Battery @ 100 €/kWh similar weight as diesel trucks, while battery trucks with similar ranges would weigh 5 to 6 tons more. Most of the fuel cell powertrain weight results from the Table 3 Assumptions for cost calculation fuel tanks, due to the number of tanks required to store the required amount of hydrogen. The second Powertrain Cost Comparison biggest contributor is the battery (currently rated at 50kWh) and the fuel cell system, based on 3 fuel cell +436% system modules (i.e., duplication of balance of plant components, piping and housings).

Looking at the future, diesel powertrains will likely +65% remain with the same weight, maybe slightly lower, given typical technology advancements but also considering additional after gas treatment components. Fuel Cell systems have the potential to Diesel Fuel Cell Battery be about 30% lighter, if one single fuel cell system FCS Battery Fuel Tank Rest Motor and E-axle E/E Cooling were to be used (therefore decreasing number of BOPs, enclosures and piping) and reduce the number of tanks by optimizing tank technology, packaging Figure 6 Fuel Cell Powertrain Cost Distribution and system efficiency. Battery powertrains also show great potential for weight reduction, with most of the The comparison shows that a diesel powertrain is still gains coming from the increase in cell energy density. the cheapest powertrain option, followed by fuel cell This increase is expected to come from the adoption system and battery being the most expensive. of solid state batteries which will have approximately 2.3 times higher energy density than current Li-Ion Furthermore, the biggest part of the fuel cell batteries. powertrain cost is attributed to the hydrogen tank system and fuel cell system. In the fuel cell system, TECHNOLOGY CHALLENGES the fuel cell stack is the largest source of cost. Its cost scales linearly with the power required. Therefore, the It is clear by now that fuel cells can be a good heavier the truck or the higher the maximum alternative as a zero emission propulsion powertrain continuous speed requirement, the more expensive also for trucks. However, the technology still needs to the fuel cell powertrain will be. be further developed to achieve the full potential described above. Among those challenges, this paper Since the fuel cell stack cost is directly related to the targeted power, the right-sizing of the fuel cell system

74 ICPC 2019 – 3.1 is, as stated above, a major challenge. The sizing Degradation Passenger HD Truck point of the fuel cell stack and system is determined Modes car 350,000 to 17,000 to by considering the required maximum continuous Stop/Go Traffic (#) power and the available heat rejection from the 400,000 34,000 Long Soak Start 6,000 to vehicle. Furthermore, the cooling capacity and the 3,000 to 4,000 required electrical power will determine at which Up (#) 13,000 Freeze Start (#) 2,000 to 2,250 1,500 efficiency point the fuel cell stack needs to run. The Hot Operation (h) 200 to 300 200 to 300 higher the required efficiency target, the larger the 5,000 to 6,000 15,000 to 30,000 Total operating necessary active area of the stack (either larger cell h or h hours (h) active area or higher cell count). Therefore, working 15 years or 10 years on cooling systems that allow the fuel cell to reject more heat is imperative to achieve lower fuel cell Table 4 Expected frequency of typical use cases system costs. with influence on fuel cell system degradation for heavy duty trucks and Another important cost driver will be related to the passenger cars daily range. Trucks have a relatively large In general, it is possible to ascertain from Table 4 that requirement for range. With current requirements of most of the use cases that typically cause achieving at least 500 km per day, the required degradation in a fuel cell will be less than or similar to amount of hydrogen would be in the range of 50 kg of passenger car vehicles. This is due to the typical use hydrogen. Today, the available tank alternatives are case (for example less urban traffic, less freeze start mostly based on the previous developments for ups, more continuous system operation) and while passenger vehicles. These tanks can store about 3 to these trucks are expected to have a major rebuild 4 kg of hydrogen. That means that the number of after the 10 years time, while cars have to last up to tanks required are large, which multiplies the number 15 years. The difference will be the number of of valves and piping linearly. This adds extra cost that operating hours, which poses a challenge to ensure would not be necessary in case of a purpose built tank that every component holds its integrity for almost 5- or usage of liquid hydrogen tanks. times more operating hours.

Durability Packaging Typical truck business powertrains are designed to Packaging a complete fuel cell powertrain (including have a durability of 1.0 to 1.5 million km. hydrogen tanks and battery) in a standard ladder- frame based truck is a challenge. Figure 8 shows the Fuel cell durability is defined in different ways. packaging space in a standard 4x2 tractor including Typically, the durability of fuel cells would mean the the possible volume that can be used to package the number of operating hours that lead to a 10% peak hydrogen tanks laterally to the frame (blue marked power degradation. This target classification has box). This vehicle configuration is the most been inherited from the automotive passenger car challenging for a fuel cell powertrain integration, as development as 10% peak power loss is a the wheel base is short and does not allow for the degradation level when most customers start to integration of long hydrogen tanks. It is assumed that notice a performance decrease during operation of the electric drivetrain is integrated in the rear axle. the car. As trucks travel frequently at their max speed Hence a powerful integrated e-axle has to be (between 80 and 90 km/h, or 60 km/h uphill), even a developed and integrated in such vehicles to small reduction in power would be noticed by truck maximize the packaging space for energy storage drivers, given the lower speeds that they also travel. systems (battery and hydrogen tanks). The packaging space for the fuel cell is shown in Figure 7 Fuel cells degrade in different ways compared to too (green marked box). The idea is that the fuel cell ICEs. For example, an important degradation should be mounted in the engine compartment to be mechnaism of fuel cells happens during start up, directly connected with the cooling loop to avoid large causing electrode corrosion. In addition, driving in a pressure drops. city in stop/go traffic also causes catalyst degradation due to voltage cycling. Hence, to understand how serious the durability problem is, one needs to understand the failure modes and how often they occur during the typical use cases. Table 4 summarizes the use cases where degradation is normally present and how frequently they are happening.

75 ICPC 2019 – 3.1

Cooling Table 5 shows the 3 main continuous vehicle speed scenarios that drive the sizing of a fuel cell system:

Vehicle Road Ambient ID Speed Inclination Temperature km/h % °C 1 90 0 35

2 80 2 35 3 50 6 35

Table 5 Sizing scenarios for a fuel cell system For each one of these requirements it is important to calculate the required power and the available cooling capacity. This drives the sizing of a fuel cell system and in turn defines its cost. If there is enough cooling capacity, then the stack active area can be minimized and hence cost can be reduced. If the cooling capacity is not enough then the fuel cell stack needs to be operated at a higher efficiency point (lower areal power density) which will drive cost higher. Figure 8 shows the calculated power for each scenario, the available cooling capacity from current cooling systems and the expected heat rejection from the fuel Figure 7 Available packaging space for fuel cell cell system with an assumed efficiency of 50% (which powertrain components would optimize the cost). As already mentioned, the hydrogen tanks will have to be mounted on the blue marked areas. AVL 500 450 estimations show that approximately 25 kg can be 400 packaged in both blue marked areas with current 700 350 bar tank technology. This amount of hydrogen will be 300 250 enough for about 250 km of range. Since this is not 200 ideal, the solution would be to install additional 150 100 hydrogen tanks behind the cabin. (kW) Capacity 50 0 Flat, 90 km/hr, 0% Hill Climb, 80 km/hr, 2%Hill Climb, 50 km/hr, 6% Current prototype vehicles from Toyota, HMC, and Incline, 35C incline, 35C incline, 35C others package their hydrogen tanks behind the Cooling Available Rejection, Heat Power, driver’s cabin. With current European truck dimension Power Heat Rejection FCS Available Cooling Capacity legislation, this would lead to a smaller cabin (no sleeper cabin anymore) or to reduced payload as the cargo volume would be reduced. On the one hand Figure 8 Power, Fuel Cell Heat Rejection and sleeper cabins are required for long-haul applications Available Cooling Capacity and on the other hand current truck body systems, like swap bodies, are optimized to use the full vehicle The first conclusion is that for the requirement driving volumes in the range of legislation. Often payload on the flat road, the currently installed cooling reductions are not acceptable or it has to take place capability of the truck is enough. However, for both in a way that standardized good sizes fits into the hill climb scenarios the available heat rejection is not body (load securing or use of standardized pallets). enough at 50% fuel cell system efficiency, which would require a resizing of the fuel cell at higher Therefore, the challenge going forward will be to efficiency point, driving the fuel cells stack costs integrate enough hydrogen on board to ensure a higher. The problem is more critical for hill climb at 50 minimum range of 500 km, and an ideal one of 1,000 km/h, as this scenario requires the cooling capacity to km avoiding hydrogen tank volumes behind the cabin. increase by a factor of 3. Research into new forms of hydrogen storage will be necessary (for example, liquid tanks) including the One might ask why cooling is such a problem for fuel required infrastructure. cells as they normally operate at higher efficiencies than diesel engines. However, there are some

76 ICPC 2019 – 3.1 reasons why cooling is an issue for fuel cells. The first is that today, fuel cells can operate at a maximum coolant temperature of 95 °C, while combustion engines typically operate between 110 °C and 120 °C. This is a limitation imposed by the ion exchange membrane, which in theory can be operated at higher temperatures, but since humidification becomes harder at higher temperatures, most systems have been designed to run at 95 °C or even at lower temperatures. As a consequence, this reduces the temperature difference between the radiator and the Figure 9 Simulation setup from AVL ambient air, which leads to less cooling capacity. The basic input for the simulation is a representative Another issue is that the air that is required for the fuel driving cycle, with respect to the usage of the vehicle cell reaction, needs to be compressed before entering and the target mileage, typically vehicle speed and the fuel cell and it reaches temperatures above what altitude over time. AVL CRUISE M allows then to map the fuel cell stack and humidifier can tolerate (max different powertrain configurations (e.g.: different no. 110 °C for humidifier membrane, air compressed can of gears, different gear steps, different battery reach up to 180 to 190 °C). This stream of air needs capacities,) to this driving cycle. The properties of to be cooled down by the same cooling loop of the each component can easily be modified to match fuel cell system which adds another heat load to the available components (off the shelf components) or to cooling loop. define the requirements of the component (if it needs to be developed). Since the air entering the fuel cell is already warmed up and will exit at 95 °C, the amount of heat leaving In the simulation setup shown above, the optimum the fuel cell system through the exhaust is also powertrain solution is achieved when the consumed negligible. energy over the driving cycle is at its minimum. (other attributes can be defined based on the individual Due to these 3 facts the cooling of a fuel cell system OEM input or use cases). poses challenges which require a new approach either through new radiator designs, more cooling Subsequently, the results of the powertrain area, higher operating temperatures or new cooling optimization are: concepts. • Fuel cell max power AVL SOLUTIONS • Battery size, evaluation of SOC • Energy consumption AVL has developed over the last years several • E-Motor power (continuous/peak) approaches, tools and methods to address the • No. of gears/gear ratios challenges discussed in the previous section. These are described in the next paragraphs. Exemplarily, Figure 10 below gives an overview how the underlying driving cycle defines the system Tools and Methods power. In this case the powertrain consists of 2 fuel cells systems (FC1 and FC2 with different sizes) The overall process of powertrain optimization is a which can be utilized according to the operating standardized task at AVL. strategy. The difference between the total power demand and the power of FC1 + FC2 indicates the Optimization of the powertrain or balancing of the power / energy which needs to be provided by the individual components respectively is done by using battery. AVL CRUISE M for vehicle simulation in combination with a DoE (Design of Experiment) approach for intelligent parameter variation as well as KPI models for optimization.

As an example, the AVL CRUISE M tool package is used for a full powertrain simulation. The picture in Figure 9 gives an overview on the simulation setup of a fuel cell powered HD application.

77 ICPC 2019 – 3.1

lose resolution on accumulation of damage for several key failure modes and drives the use of transient capable models. The AVL developed models that can account for this different time constants and the dynamics and allow one to generate hydration profiles in the cell as show in Figure 11.

(a)

Figure 10 Fuel Cell and Battery power balancing

Model Based Approach for Durability (b) As with performance modeling, a high-level powertrain model is used to understand the implications of any energy storage element and the hybridization strategy on how vehicle drive cycles Cell Relative Humidity map to actual stack cycles. Figure 11 Cell hydration profile and the estimated Physics-based models of failure modes and stressors durability (a) high temperature, low permit creation of robust, accelerated lifetime test hydration, (b) target temperature and methods which can be linked through models to good hydration actual stack durability estimates in vehicle simulations. These models are also critical to The integration of such physical degradation models properly cascading verifiable requirements to the of the fuel cell system and battery in the powertrain subcomponents of the membrane electrode models enables an optimum sizing of the components assemblies, like the membrane, or cathode catalyst as the lifetime degradation of the most costly powder type and loading. powertrain components is considered.

Physical models also allow development of transfer Modular Fuel Cell Systems and Operating functions for performance degradation cycling tests Strategies for subcomponents that are often done in test cells Current fuel cell demonstration trucks make use of with minimal in-plane gradients, whereas the in-plane fuel cell systems that were designed and developed gradients present in a full-scale automotive cell cause for passenger car applications or even for stationary complexity in both how damage is accumulated, as applications. However, the different needs of well as how it manifests in the overall device commercial vehicles must be acknowledged and thus performance. dedicated fuel cell systems, considering power output, operating hours, lifetime, and reliability need Lastly, as many of the stressors tend to be transient to be developed. in nature, and since there is a wide range of time constants of response of the stack to condition While current passenger FCEVs share a common changes, dynamic models are needed to understand targeted fuel cell power that ranges around 100 kW, the stack degradation response to system inputs. For commercial vehicles possess a large variety of use instance, in response to parameters like current draw, cases and vehicle weight classes with different power air flow or air pressure, the performance of a stack will requirements. The rationale of developing dedicated change and stabilize within seconds. However, in fuel cell systems to high power demands remains response to changes to inlet hydration, the time questionable as the applicability to different use constant of response of relevant parameters to cases (long-haul, regional and last-mile delivery) lifetime (i.e. membrane hydration) can be on the order might not be given. By taking advantage of an of minutes. This causes the use of static models to innovative modular fuel cell system approach,

78 ICPC 2019 – 3.1 consisting of multiple smaller fuel cell systems, (see Case 1 in Figure 12). It is better to run fuel cell maximum flexibility to meet requirements of different systems at mid-load where efficiency is high, thus not commercial use cases can be achieved. As such, too much heat is produced, and the cell voltages are modular fuel cell systems can be applied to different in a range that does not cause degradation. Systems vehicle classes, e.g. 7.5 t, 18 t, and 40 t classes, that are not needed to fulfill the power request of the depending on the number of modules installed vehicle should be shut down completely and only onboard. This approach requires a better turned on as more power would be requested. In understanding of the trade offs of how to operate the order to achieve a high overall lifetime, the several stacks to maximize system performance, operational hours should be distributed among all fuel durability and robustness. Figure 12 shows an cell systems of the truck powertrain. example of 2 cases in which the total power draw from 3 modules is the same, but different variations. To achieve the required performance, efficiency, and operational lifetime in commercial vehicle applications, sophisticated monitoring and control concepts are required to mitigate degradation of the fuel cell system. Typically, the fuel cell hybrid system is comprised of the fuel cell itself as well as additional energy-storage devices (such as HV batteries and super/ultra-capacitors). The primary control goal is to meet the required power demand (i.e. the requested mechanical power) at all times, while operating the fuel cell and energy-storage devices within their optimal regions. The energy-storage devices are charged via recovered braking energy (recuperation) and via the fuel cell system when the overall power

demand is low. Accordingly, these components are Figure 12 Strategies of power distribution between used to supply peak power and represent a long-term the systems energy buffer. Besides the mechanical/driving power, major electric loads may also occur from vehicle Table 6 summarizes the strategy for each system and auxiliaries like refrigeration units (in refrigerated cargo the expected impact on durability or performance. vehicles), which have a substantial influence on the overall powertrain and fuel cell system operating strategy. CASE 1 CASE 2 FCS Full power High power Due to the variability of possible system 1 High heat Balanced heat configurations and dimensions in HD vehicles production production regarding energy storage-devices and electric loads, Potential Cell voltage at non- modularity is the key to foster efficiency and lifetime degradation due degrading level improvements of the interconnected powertrain to hot operation elements. Balancing of loads over multiple fuel cell FCS Low power High power systems will result in maximizing the energy efficiency 2 Degradation due Balanced heat of the overall fuel cell powertrain by reducing the load to high cell production that is applied onto each single system. Obviously, voltage Cell voltage at non- maximum powertrain efficiency further increases the degrading level driving range of the truck and keeps the heat rejection FCS Minimum load to Idle power of the fuel cell systems at a reasonable value, 3 avoid OCV Degradation due to simplifying the cooling system. At the same time, very high cell voltage durability- and lifetime-impairing operating conditions (close to OCV) can be avoided by maintaining preferred cell voltages in each stack. Table 6 Comparison of 2 different power distribution strategies While balancing of electric loads over several fuel cell systems is an important measure to reduce stack degradation, additional levers need to be triggered to Even though multiple smaller fuel cell systems achieve targeted operational lifetimes in commercial provide a high flexibility regarding operational vehicle applications. As such, reduction of system strategies, it is inefficient to operate single fuel cell dynamics is a key solution. systems at high power while operating others at low power or even idle power considering fuel consumption, heat rejection and durability/lifetime

79 ICPC 2019 – 3.1

actual state-of-health of the stacks while calculating the setpoints for each actuator. Typically, the state- of-health is determined by monitoring the voltage of each single cell in the stacks. A deviation from the expected cell voltage would indicate that the cell or cells are operated in an undesirable condition, which could result in accelerated degradation of the stack. However, the voltage value does not provide insight into the underlying cause of the problem, thus a dedicated countermeasure cannot be taken, and a series of potential measures need to be performed following a certain reaction matrix. The application of such a cell voltage monitoring (CVM) remains questionable, considering the high number of sensors (>400 per stack), the associated cost and low quality of information.

In contrast, AVL developed a methodology, called THDA (Total Harmonic Distortion Analysis) that allows to distinguish between different root-causes of degradation-triggering operating conditions, while at the same time reducing the number of sensors to two per stack. Despite the low number of sensors, AVL’s methodology and algorithms are powerful enough to still achieve single cell resolution, providing comprehensive state-of-health information from single cell to stack level. THDA detects whether the stack is operated in too dry or too wet conditions, resulting in membrane dry-out or flooding and whether there is a starvation of reactants. This additional information on the root-cause enables targeted countermeasures and thus reduces the time in which the stack is operated in lifetime-impairing conditions, thus prolonging the overall durability of the fuel cell powertrain. Since the methodology can be implemented directly into the fuel cell controls and the fuel cell DC/DC boost converter, no additional hardware cost is added to the fuel cell powertrain, Figure 13 Examples of dynamic operating thus achieving a significant cost reduction compared strategies for modular fuel cell systems to the current state-of-the-art cell voltage monitoring.

Assuming a dynamic power request of 125 kW/s, as E/E Integration shown in Figure 13, it is obvious that a single high- power fuel cell system will need to bear the full load As the power levels required for HD trucks are high, change by itself (Figure 13a). In contrast, in a fuel cell the usage of high voltage systems is normally truck powertrain consisting of multiple smaller fuel cell preferred. This means that the connection of several system modules, lower dynamic load changes need fuel cell systems and batteries to the inverters and to be borne on system level while achieving the same motor needs to be done with different DC-DC overall 125 kW/s power request on powertrain level converters. AVL has investigated several DC-DC (see Figure 13b and Figure 13c). At lower power converters solutions and created several trade off demands, single fuel cell systems in a modular fuel models that can be used. cell powertrain could even be turned off, reducing the overall number of operating hours of each system, thus potentially prolonging the overall powertrain durability and lifetime (Figure 13c).

Real Time On-Board Diagnostics In order to achieve highest durability of the fuel cell stacks, the operating strategies need to consider the

80 ICPC 2019 – 3.1

Figure 14 shows the potential DC-DC converter REFERENCES configurations that could be used to connect the fuel cell systems. Each one of them has different trade [1] FCH Europe, 2017. Development of Business offs that need to be observed based on the customer Cases for Fuel Cells and Hydrogen Applications requirements. for Regions and Cities - FCH Heavy-duty trucks. Online: http://www.fch.europa.eu/sites/default/files/1711 21_FCH2JU_Application- Package_WG1_Heavy%20duty%20trucks%20% 28ID%202910560%29%20%28ID%202911646 %29.pdf, accessed on March 29, 2018 Figure 14 Potential DC-DC converter wiring [2] AVL: Hybrid fuel cell powertrain. In: Electric & Hybrid Vehicle Technology International 01/2018, CONCLUSION S. 176 [3] Schenk, A, Berg, F, 2018, 'Konzeptfahrzeug mit Fuel cells can offer several advantages as a zero Brennstoffzellen-Plug-In-Hybrid', ATZ 10/2018, emission powertrain for heavy duty trucks. These pp. 30-35 advantages are related to the quicker refilling times, [4] Rechberger, J, Schenk, A, 2017, 'Auslegung der low weight and low total cost of ownership. However, Medienversorgung für ein automobiles the technology still has several challenges in a truck Brennstoffzellen-System’ Ladungswechsel im vehicle with regard to cost, durability, cooling and Verbrennungsmotor 2017 - 10. MTZ-Fachtagung packaging. AVL is engaged in solving these issues by Proceedings. Springer Vieweg, Wiesbaden, pp. offering innovative solutions like modular fuel cell 1-12, DOI: 10.1007/978-3-658-22671-8_6 systems, advanced operating strategies, high voltage [5] Berg, F, Schenk, A, 2018 ‘ PEM fuel cell boost converters with integrated real-time on-board powertrain – which application makes sense – a diagnostics as well as all necessary simulation and study’ Eco-Mobility 2018, Vienna, November 13, testing environments. AVL will continue to work and 2019 advance the technology focusing efforts in the years to come on the hydrogen storage, fuel cell system cost and cooling issues, so that weight and cost can DEFINITIONS, ACRONYMS, be addressed and reach the targets set forth by the ABBREVIATIONS industry. OEM Original Equipment Manufacturer. ACKNOWLEDGMENTS TCO Total Cost of Ownership The authors would like to thank all contributors to this ICE Internal Combustion Engine paper from AVL Fuel Cell Canada (Amy Nelson, DOE Design of Experiments Roger Penn, Sascha Mielke) as well as from AVL SOC State of Charge Graz (Christian Niedermayr, Armin Traussnig, Franz FCEV Fuel Cell Electric Vehicle Hofer) and AVL Steyr (Wolfgang Gruber, Michael Kordon). OCV Open Circuit Voltage HV High Voltage HD Heavy Duty CVM Cell Voltage Monitoring THDA Total Harmonic Distortion Analysis

81

ICPC 2019 – 3.2 Commercial vehicle battery solutions

Krzysztof Paciura Cummins Inc. Copyright © 2019 AVL List GmbH, Cummins Inc.and SAE International

ABSTRACT generators from 0.6 kVA to 20,000 kVA under the MARKON®, STAMFORD® and AvK® brands. This paper describes the research and development journey that Cummins Inc. is taking to secure a High-speed electromechanical machines: Expertise market-leading position in electrified powertrains for is assured through Cummins Turbo Technologies and commercial vehicles. Accelerating the migration from Holset, who are one of the world’s largest existing internal combustion engines (ICE), these turbocharger manufacturers with annual sales circa developments underpin the next-generation of more- $1 billion, having been integrated into the Cummins electric propulsion solutions, including full family of brands in 1973. electrification and hybrids using efficiency-optimised thermal propulsion. As well as summarising all of A changing world for propulsion these activities, here Cummins introduce the However, whilst the diesel engine has revolutionised development of a novel approach to high power transportation and industry, particularly in medium density electric traction based on Synchronous and heavy-duty applications, the landscape for this Reluctance Permanent Magnet Assisted electrical success is changing. It is now known that trucks used machines, and discuss integration of power for regional and urban logistics cause up to 45% of electronics and other electrification accessories. air-pollution in and around built-up areas, with devastating environmental and socio-economic INTRODUCTION impacts. The UK alongside many other countries targets migration to a zero-emission vehicle fleet over Who are Cummins Inc.? the coming decades, but a significant number of cities including London, Paris and Barcelona have Cummins are the world's leading independent commited to blanket diesel-bans far sooner, producer of diesel engines for commercial fundamentally risking logistics chains. applications, ranging from 55 to 3,500 horsepower. As well as diesels, Cummins currently design, manufacture and supply a complete line of natural Battery electric vehicles (BEVs) remain the most gas engines for many on-highway and off-highway credible approach to address this challenge, but markets, including heavy-duty and medium-duty whilst the broader evolution is occuring incrementally trucks, buses and plant equipment. Additionally, alongside both technology-driven and organic cost Cummins supply engines into multiple industrial and reductions, the immediate need for cost-effective power generation applications including agriculture, commoditised medium and heavy duty BEV construction, mining, marine and military equipment. powertrains remains unmet. In the meantime, the role of efficient thermal propulsion and hybrid technology is vital to realise rapid reductions in CO2 emissions. Cummins have realised significant vertical integration in the supply chain, and also develop and supply value-added auxiliary components and subsystems Cummins’ journey towards electrification across the electric machine, electromechanical Cummins aim to be at the forefront of this new electric machine and electronic system domains. Cummins powertrain revolution, with a technology-driven can subsequently leverage globally-leading expertise approach to secure a position as the world’s leading and joint development partnerships with key provider of end-to-end electrification solutions in the stakeholders within the group as follows. on and off-highway sectors.

Electric machines: Cummins Generator Technologies Cummins established a dedicated Electrification have proven prominence globally, and currently Team in 2017 to support their Research and manufacture the world's broadest range of AC Technology ambitions in this space, which are central

82 ICPC 2019 – 3.2 to their overall global product roadmaps. The Electrification Team ensure that Cummins continue to make the best possible strategic decisions in terms of internal investments in component and system technologies, building the new capabilities needed, and stimulating the external partnerships to meet their customers' needs in these prioritised markets.

Cummins have already gained significant expertise in powertrain electrification, and have long been the largest supplier of engines for hybrid commercial vehicles. Targeting the progression to zero emissions, over the last 15 years Cummins’ global Research and Technology team has been privately and publically developing a range of diverse electrification technologies, which can be applied directly to improve combustion engine efficiency and CO2 performance, as well as enable hybridisation of Figure 1 High-speed electrical machines for the ICE-based powertrains and full electrification in the more-electric engine. future. ELECTRIFICATION OF WASTE HEAT RECOVERY During this period Cummins have successfully These approaches have been further explored to completed many high-profile projects from concept- level upwards, actively partnering with legislators, realise novel architectures for electrically-assisted suppliers and customers to realise new more-electric turbocharging, realised by integrating new electrical powertrain solutions. machines with existing high-speed machinery. Cummins are also implementing thermal heat recovery mechanisms for further efficiency gains. Maintaining vertical-integration of the components with the largest impact on performance, quality and system power, Cummins have also recently acquired Brammo and Johnson Matthey Battery Systems, to ensure strong capability in energy storage as part of the end-to-end electric powertrain offering.

CUMMINS CURRENT ELECTRIC POWERTRAIN PORTFOLIO

Cummins’ vision to penetrate the growing market for electrification technologies is evidenced by the commercial entry of a first generation of these products, summarised as follows.

Electric machines Figure 2 Electrically-assisted turbocharger design. Focussed initially on high-speed electromechanical machines, Cummins have developed a number of HYBRID TECHNOLOGIES proprietary technologies for energy recovery [1]. Cummins have scaled these developments to realise a highly compact but powerful permanent magnet E-COMPOUNDING ENERGY RECOVERY motor topology. This can be used for direct drive Cummins have developed and integrated a suite of propulsion in pure BEV or range-extender solutions high-speed electromechanical machines which for early adopters, alongside a downsized 2.8L EURO enable energy recovery from exhaust gases, as well 6 engine, or in hybrid stop-start applications. as enhancing efficiency through electrically-assisted compression of the inlet and recovered airflows to the combustion engine.

83 ICPC 2019 – 3.2

THE TECHNICAL CHALLENGE OF FULL ELECTRIFICATION

Electric machines The drive-cycles and diverse payload conditions of medium and heavy duty trucks place extreme demands on the powertrain performance in terms of: startability (peak torque at wheels above 20,000Nm – 2710Nm at differential, using standard trans-axle ratio); gradeability (continuous torque at differential 1,000Nm); speed (0-3000RPM at differential at 2710Nm); and corresponding power (>200kW). High powertrain efficiency over the whole drive cycle is

also a vital consideration for mainstream adoption. Figure 3 Hybrid powertrain integration for a range- extender electric vehicle. Whilst only partially meeting these torque and speed requirements, the existing state-of-the-art BEV Battery Packs powertrain architectures based on synchronous Through strategic growth and partnerships, Cummins motors commonly contain up to 100kg of rare-earth have a full range of proprietary battery technology, to permanent magnets as the primary excitation support energy recovery and plug-in electrification. medium to generate the very high fields required. As well as restricting the cost, weight, size and efficiency over the drive-cycles targeted, dependence on highly volatile mineral resource supplies of these exotic materials fundamentally prevents commoditised volume manufacture, and capability to meet future electrification demands.

Power electronics integration Electric powertrains must also compete with ICE architectures in terms of vehicle-level and system- level integration. ICE power units are already highly integrated, having matured over many product generations to include all of the key auxilliary functions into a single warrantied product.

In contrast, this level of integration is yet to be realised for BEV powertrains, for which the primary challenge faced by technology developers of key subsystems and components – such as power electronics and controls – has been realising the baseline functionality to achieve the operational performance Figure 4 Cummins’ modular pack architecture. requirements. The integration challenge extends far beyond the interfaces and electro-mechanical Modules: Focussing on individual customers’ unique aspects, and thermal management of electronic operations and needs, Cummins’ high energy density components becomes critical given the temperatures modules underpin significant scalability and flexible expected as they approach their limits to deliver the vehicle-level integration. extreme performance targeted.

Enclosure: Cummins have applied advanced ELITS: EFFICIENT LIGHT INTEGRATED composite materials to realise a super-lightweight TRACTION SYSTEM case, increasing energy density and durability. In response, Cummins – working closely with the Battery Management System: Cummins have University of Nottingham’s Power Electronics, developed proprietary control architectures to Machines and Control Group – are developing totally maintain safe operation and isothermal management new high power architectures for medium and heavy under extreme operating conditions. duty BEV powertrains, which are highly integrated and require no rare-earth permanent magnets.

84 ICPC 2019 – 3.2

Cummins’ approach of delivering extreme startability (20,000Nm at the wheels) and 10,000Nm continuous torque at speed. Using extreme gearing to radically reduce the motor peak-torque requirement, Cummins aim to deliver a Power electronics breakthrough in power density by uniquely targeting an increase in electric machine speed within this The development of proprietary power electronics will context. This is being developed alongside direct- focus on the intelligent control of the rotor to support integration and cooling of power electronics into the safe and reliable operation at the motor speeds traction motor housing, enabling intelligent control in targeted, as well as shared cooling with the electric a single or modular motor configuration. machine to enable direct integration without the addition of further auxiliary components.

Modularity and scalability Interfaces for plug-and-play modularity are critical to scale the technology for heavy-duty applications, including the wide range of off-highway market sectors of strategic importance to Cummins.

INNOVATION OUTPUTS The design-performance of these topologies corresponds to a 5-fold increase in maximum efficient motor speed in a >200kW motor, for which reluctance excitation alone is insufficient. However, the novel stator and rotor design is shown (by simulation and Figure 5 Power-speed and torque-speed curves. prototyping) to achieve the characteristic rpm.√kW greater than 200,000, which reduces the generation With this industrial-academic partnership yielding a torque requirement of permanent magnets from number of novel high-speed reluctance machine 70%→10%. With careful engineering of the magnet topologies, Cummins are now focused on scaling and properties and integration, this breakthrough enables integrating disruptive synchronous reluctance motors low-cost ferrite material to replace rare-earth into scalable powertrain systems, which only require magnets, catalysing a 100-fold reduction in raw minimal permanent magnet assistance. material costs, and an unconditional supply chain.

The work within the ELITS programme therefore yields the first cost-effective alternative to permanent magnet synchronous-motors for medium and heavy duty BEV powertrains. Innovations address the specific technical challenges of safely and reliably achieving these motor speeds.

STATOR ELITS leverages new electrical-material development (6.5% Si-steel), lamination engineering, distributed- winding geometries and advanced cooling.

Figure 6 Magnetic field distribution at high speed. ROTOR Embodied by high-strength materials, lamination and Electric machine performance domain geometries optimising reluctance-excitation, ferrite integration and direct cooling. The step-change in maximum revolutions-per-minute (15,000 RPM) in a motor of this size and cost is fundamental to Cummins’ capability to uniquely deliver efficient power up to 220kW over an unprecedented range of motor speeds, using only very low permanent magnetic fields. Transmission development Also central to the solution is the development of an integrated multi-speed transmission, which is capable

85 ICPC 2019 – 3.2

REFERENCES

[1] Gerada, D et al. (2013), High-Speed Electrical Machines: Technologies, Trends, and Developments. Members, IEEE. Volume: 61 , Issue: 6 , June 2014. [2] Walker, A et al. (2016), Development and Design of a High Performance Traction Machine for the FreedomCar 2020 Traction Machine Targets, Members, IEEE. [3] Nardo, M et al (2016), Comparison of Multi- physics Optimization Methods for High Speed Synchrnous Reluctance Machines. Members, IEEE. Cummins Generator Technologies, Stamford, UK

[4] Gerada,D et al. (2011), Electrical Machines for Figure 7 Rotor geometry parameter space [2]. High Speed Applications with a Wide Constant- Power Region Requirement. [5] Erik Odvarka, E et al. (2009), Electric Motor- Generator for a Hybrid Electric Vehicle, Journal of Engineering Mechanics.

Figure 8 Reluctance torque based on the distribution of total barrier thickness [2].

CONCLUSION

Cummins have invested heavily in electrification technologies over the last decade, focussing on the critical components of electrified powertrains and leveraging the unique skillset across Cummins’ employee base, including mechanical, thermal, rotor dynamics, electro-magnetics and power electronics engineers. As a result, Cummins is ideally positioned to become a leader in the electrification value chain as a global Tier 1 supplier. The technological innovations stimulated by the ELITS project give Cummins a competitive advantage in electrical machine design and power electronics development, specifically targeting medium to heavy duty powertrains in sectors which are key to reducing the environmental impact of diesel transport, in line with global government-level ambitions and legislation.

86

ICPC 2019 – 3.3 The smart eCVT hybrid system for 2020+ Commercial vehicle application

Dr.Zhiqiang Lin, Song Zhang, Tao Chen, Zhengsong Mao Guangxi Yuchai Machinery Co., LTD

Dr. Christoph Schoerghuber AVL Commercial Driveline & Tractor Engineering GmbH, Steyr

Copyright © 2019 AVL List GmbH, John Deere and SAE International

ABSTRACT next to the achievement of highest fuel consumption benefits, one of the main targets for Yuchai is the In this paper a new e-CVT hybrid system for various flexibility to apply the transmission to different commercial vehicle applications is presented. This commercial vehicle applications. Target applications new hybrid system is combining the power sources of of the developed hybrid powertrain are city buses but a combustion engine with two high-speed e-motors also highway coaches and medium duty trucks such which are highly integrated into the e-CVT as concrete mixers, cranes or delivery trucks with a transmission. With the help of a new transmission gross vehicle weight between 16 and 20 tons. Thus, concept including different operating modes, high the powertrain must achieve high system efficiencies efficiencies combined with comparatively low product for city as well as overland and highway driving and cost are achieved. The high efficiency of the must be adaptable to different maximum vehicle powertrain and the perfectly coordinated interaction velocity and gradeability requirements. Also, a PTO of the three power sources ensure high fuel savings unit is required for some of these applications. without compromising drivability and driving comfort. This is achieved, in particular with dynamic load AVL developed for Yuchai a new hybrid powertrain profiles as they are usual for city bus applications, via system which exactly fulfills these demands. Via power-split and pure electric vehicle operation mode. electric power split with continuous variable In addition, an overdrive operating mode enables transmission (CVT) property and further transmission efficient driving at constant high vehicle velocities for modes this smart e-CVT hybrid system enables highway applications. Thereby, an optimum operating lowest fuel consumption for different commercial range for the combustion engine is achieved via a vehicle applications. Due to its modular and mechanical power flow and parallel hybrid operation. comparatively simple design the hybrid powertrain The operating mode is selected on the basis of a well- can be adapted to different applications with developed hybrid strategy. Furthermore, the hybrid moderate production costs. system can be easily adapted to different commercial vehicle applications. The optional equipment with a MECHANICAL LAYOUT OF THE E-CVT PTO module with pure electrical and mechanical HYBRID SYSTEM operation makes the e-CVT system also suitable for many truck applications. The powertrain consists of the engine (ICE), 2 electric motors (e-motors), a battery, a dry clutch, the e-CVT INTRODUCTION and the rear axle. The e-CVT consists of the input shafts for the e-motors and the engine, gear boxes for The development of highly efficient commercial speed reduction of the e-motors, a mode selector vehicle powertrains is subject of ongoing research unit, 2 planetary gear sets including (hollow) shafts, and development activities all over the world. the transmission output shaft as well as a limp-home Reduction of fuel consumption is the main driver for and PTO unit. hybridization of commercial vehicle powertrains. But

87 ICPC 2019 – 3.3

Figure 1 Renderings of the designed e-CVT hybrid system

In Figure 1 renderings of the finally designed e-CVT operated without any torque interruption leading to hybrid system can be seen.The left-hand side picture highest driving comfort. Due to the battery as energy shows the e-CVT transmission with the two e-motors storage device the electric motor and generator are and the dry clutch actuation including a view into the almost independent from each other and can be gear box. The powertrain including engine, dry clutch controlled separately. For highly dynamic drive cycles and e-CVT with both e-motors but without rear axle high fuel consumption benefits are achievable also and battery can be seen on the right-hand side. The because the engine can be always operated at best e-motor 1 (EM1) is connected to the sun gear of fuel consumption point, independently from vehicle planetary gear set 1 (PG1). In between there is a gear velocity. But for driving at constant (high) speeds the box with fixed gear ratio. The ICE is connected to the electric efficiency chain from generator to motor gets dry clutch which then is connected to the carrier of quite bad resulting in higher fuel consumption. PG1. The e-motor 2 (EM2) is connected to the sun Furthermore, the mechanical layout requires high gear of planetary gear set 2 (PG2). In between there power for engine, electric generator and motor and is a shift able gear box with 2 gears as well as a thus product costs are quite high. neutral position. The carrier of planetary gear set 2 (PG2) is fixed to the housing of the e-CVT. The rings Parallel hybrids require only one electric machine with of both planetary gear sets are connected to the moderate power and the mechanical layout is quite transmission output shaft. Thus, there is a fixed ratio similar to conventional powertrains which has positive between sun and ring gear at PG2, and EM2 is effects on product and integration costs. High system always connected to the transmission output shaft, efficiencies are achievable also for driving at constant since a gear is engaged at EM2 gear box. PG1 is high speeds because of the mechanic power flow and acting as power split or summation device enabling a the possibility of load point shifting of the engine or CVT operation of the transmission. boosting. However, the engine cannot be operated at best fuel consumption point continuously because Furthermore, the transmission consists of a mode engine speed and vehicle velocity are always selector unit where either the sun gear or the carrier coupled, and their relation is depending on the of PG1 can be blocked to the housing. Thus, selected gear. Thus, for city and overland driving additionally to the CVT-mode a pure electric vehicle highest fuel consumption benefits cannot be achieved mode (EV-mode) and a direct ICE mode with with parallel hybrids. Furthermore, there are torque overdrive gear ratio (OD-mode) can be selected. The interruptions during gear shifting of transmission different transmission modes are chosen very which are affecting the driving comfort of the vehicle carefully. Therefore, different hybrid and electric negatively. powertrains are investigated in a first step. Battery electric vehicles require only one electric Investigations on standard layouts of electrified machine with full power and can be operated in zero powertrains emission zones. For medium and heavy-duty commercial vehicle applications they are typically Studies on serial and parallel hybrids as well as pure equipped with shift able transmissions to increase electric vehicles form the basis for the development gradeability for vehicle start-up. Thus, there are of the transmission modes. Compared to parallel torque interruptions during gear shifting of hybrids, serial hybrids are preferable due to high transmission which are affecting the driving comfort driving comfort and comparatively simple mechanical of the vehicle negatively. Furthermore, electric topology and control strategy. Serial hybrids do not require shift able transmission and thus can be

88 ICPC 2019 – 3.3 vehicles require big battery capacities to achieve system is equipped with different transmission modes acceptable driving range. enabling the powertrain to be operated as power-split hybrid, parallel hybrid and pure electric vehicle. In Properties of the developed transmission modes Figure 2 the layouts of serial and parallel hybrids as well as electric vehicle powertrains are compared to To combine all the advantages of the described the different modes of the e-CVT hybrid system. electrified powertrains the developed e-CVT hybrid

Figure 2 Layout comparison of serial hybrids, parallel hybrids and electric vehicle powertrains with the different modes of the developed e-CVT hybrid system

The e-CVT hybrid system in electric power split EXPLANATIONS ON THE E-CVT HYBRID operation mode (CVT-mode) allows the ICE always SYSTEM MODES operating at best fuel consumption point resulting in highest powertrain efficiencies as well as driving EV-mode comfort. The resulting properties are similar to those of serial hybrids with the advantage of an additional If EV-mode is selected the carrier of PG1 and thus the mechanical power flow, leading to a further increase engine shaft is blocked to the housing of the e-CVT of efficiency. Via OD-mode the same properties as for hybrid system. Similar to the connection of EM2, EM1 parallel hybrids can be achieved resulting in highest is always connected to the transmission output shaft efficiency values even for highway driving. Via EV- with a fixed gear ratio. Because of limited battery mode the powertrain can be operated with zero capacity the EV-mode is typically used for low vehicle emissions and recuperated battery energy can be velocities. The energy for vehicle propulsion is utilized for vehicle propulsion. provided by the battery to both e-motors. The requested transmission output power in general is Due to the combination of electric and mechanic provided by both e-motors as sum of the individual power flow enabled via the different modes the overall power values for vehicle propulsion and recuperation. system power is resulting as a combination of ICE and e-motor power individual power values of the OD-mode different power sources can be reduced compared to If the sun gear of PG1 is connected to the housing, e.g. serial hybrids, leading to lower production costs. EM1 input shaft is blocked and thus EM1 is not used. Thus, with the help of the new developed e-CVT In this so-called OD-mode there is a fixed ratio hybrid system the advantages of serial and parallel between ICE and transmission output shaft with an hybrids as well as electric vehicles are combined. overdrive gear ratio. Due to the resulting engine operating points at small engine speed regions for high vehicle velocities the OD-mode is enabling high

89 ICPC 2019 – 3.3 system efficiencies for highway driving. If EM2 gear FUEL CONSUMPTION SIMULATION OF box is switched to neutral the vehicle is driven by the A CITY BUS WITH E-CVT HYBRID ICE only. If a gear is selected at EM2 gear box, additionally EM2 is connected to the transmission SYSTEM output shaft and the powertrain is operated as parallel hybrid. The OD-mode can only be used for higher For estimation of fuel consumption benefits which are vehicle velocities because the engine’s idle speed is achievable with the e-CVT hybrid system an overall limiting the minimum vehicle velocity. The requested vehicle simulation of an 18-ton city bus is performed. transmission output power in general is provided by The vehicle simulation is done with AVL Cruise based the ICE and EM2 as sum of the individual power on longitudinal dynamics of the vehicle. The values. Thus, this mode can be used for both vehicle powertrain model is split into engine (including 0,5 kW propulsion and recuperation. constant mechanic auxiliaries), e-CVT (including e- motors), battery (including battery management and CVT-mode 1 kW constant electric auxiliaries) and rear axle. Additional to the mechanical parts of the hybrid If neither PG1 sun gear nor PG1 carrier are blocked, powertrain also the transmission controller (TCU) and the e-CVT hybrid system is operated in CVT-mode as the hybrid controller (HCU) must be considered within continuous variable transmission, i.e. the ratio vehicle simulation. The e-CVT and the required between transmission output and transmission input controllers are modelled in Simulink and implemented shaft can be adjusted between 0 and a certain finite as dynamic link libraries into the powertrain model in value continuously. The requested transmission AVL Cruise. output power is provided by the ICE, EM1 and EM2 as sum of the individual power values. Because of the To perform a fuel consumption comparison an power split of mechanical to electrical and back to additional simulation of the same city bus with mechanical power, one e-motor typically is operated identical engine (including mechanic auxiliaries) and as generator and one e-motor as motor. Thereby, rear axle but with automatic transmission (AT) and PG1 is acting as power split and summation device. torque converter is performed. Thereby, a state-of- In this CVT-mode it is possible to operate the ICE at the-art AT transmission, a torque converter with lock- best fuel consumption point and to choose the split up clutch and an engine with start-stop function are between ICE power and battery power provided to the modelled. Both Cruise vehicle models can be seen in e-motors within certain limits for further efficiency Figure 3. optimization.

Figure 3 AVL Cruise vehicle simulation of a city bus. Left hand side: Conventional city bus powertrain with torque converter and AT; Right hand side: city bus powertrain with DHT.

For fuel consumption comparison, both vehicles are (CCBC). The resulting engine operating points for simulated driving on the Chinese city bus cycle both powertrain models are depicted in Figure 4.

90 ICPC 2019 – 3.3

Figure 4 Engine operating points of a city bus vehicle simulation driving on CCBC drive cycle. Left hand side: City bus powertrain with torque converter and AT; Right hand side: City bus powertrain with DHT.

As can be seen the engine operating points of the paper. The carefully developed gear arrangement conventional powertrain are located in a big region results in positive properties for different commercial below 1500 rpm. Whereas the engine operating vehicle applications. The mechanical layout is points of the powertrain with e-CVT (in CVT-mode) comparatively simple and offers the use of high- are located at the best point curve, connecting the speed e-motors, resulting in optimum packaging and best fuel consumption points depending on engine low product costs. With the help of a mode selector power. The resulting fuel consumption results are unit the new developed e-CVT hybrid system depicted in Figure 5. combines and improves the advantages of serial and parallel hybrids as well as pure electric vehicle powertrains. Via a power split operation mode with continuous variable transmission function (CVT- mode) the engine can be operated at best fuel consumption point. In addition, an engine direct drive mode with overdrive gear ratio and parallel hybrid function (OD-mode) allows the engine to operate in areas with highest efficiency, resulting in lowest fuel consumption even when driving on highway cycles. Furthermore, an electric vehicle mode (EV-mode) is used to drive with zero emission and to utilize the recuperated battery power. The simulation results show the big benefits in terms of fuel consumption compared to conventional state-of-the-art Figure 5 Simulated fuel consumption results of an powertrains, i.e. the fuel consumption can be reduced 18t city bus driving on CCBC drive cycle by 38% with the help of the e-CVT hybrid system. with conventional and DHT powertrain. Optionally the e-CVT system can be equipped with a The fuel consumption results of the powertrain with e- limp-home and a PTO module. The limp-home CVT are based on a balanced state of charge of the module ensures a safe trip with reduced functionality battery where battery energy at beginning and end of even in electric failure mode of the e-CVT hybrid the simulation is identical. With the e-CVT hybrid system. The optional equipment with a PTO module system a fuel consumption benefit of about 38% can with pure electrical and mechanical operation makes be achieved. This benefit is a result of operating the the e-CVT system also suitable for many truck engine at best point curve, recuperation of braking applications. power, engine start-stop and driving purely electrically utilizing the recuperated battery power. REFERENCES

SUMMARY AND CONCLUSION [1] EP-0967102A2 [2] CN-203766482U A new hybrid powertrain with e-CVT transmission for [3] Müller, H.: Die Umlaufgetriebe. Springer 1998 commercial vehicle applications was presented in this

91

ICPC 2019 – 3.4 Tractor/implement systems – the next generation

Dr.-Ing. Joachim Sobotzik John Deere European Technology Innovation Center Kaiserslautern, Germany

Craig Puetz John Deere Global Tractor Engineering, Waterloo, IA, USA

Copyright © 2019 AVL List GmbH, Guangxi Yuchai Machinery Co., LTD and SAE International

ABSTRACT architectures, not practicable with traditional approaches. Electric dive systems are applied in vehicles and mobile machinery since decades, wherever the The current work is built on merging of the 2nd and emission-free operation was required. E.g. in the 3rd category to create a next generation of material handling industry, electric drives represent agricultural systems, highly integrated tractor- the state-of-the-art, being integrated in a variety of implement propulsion systems. system configurations, from basic operator controlled battery-electric systems to hybrid architectures where Tractors and implement designs have shown partial battery electric operation, the availability of an significant progress in the last decades. Anyway, onboard power generation system and sophisticated getting closer to the performance and productivity of power controls can be integrated. highly optimized and specialized self-propelled machinery, tractor/implement systems continuously In passenger cars and commercial on road vehicles, increased in size and weight. One cause for less- hybrid power trains, consisting of an electric drive than-optimum systems has been the independent system, battery energy storage and a combustion optimization of tractors and implements, gains engine have gained a noticeable market share, possible thru integrated systems were not realized. particularly since the exhaust gas after treatment for diesel engines required significantly increased efforts A key advantage coming with distributing traction driven by the latest emission regulations. forces from the tractor to implements, using traction assist axles, has been demonstrated and realized in In precedent publications [1] three categories of the past. Less tractor weight and improved payload electric drive application in off road machinery were on implements were initial targets. Managing the discussed: efficiency of the overall system, tractor, implement and process is expected to drive additional Engine and vehicle auxiliary drives, e. g. radiator improvements. fans, coolant pumps, AC-compressor drives, substitutes for alternators. The paper will describe test results and expected improvements coming with a tractor-centric efficiency Power transfer from a mobile power source like a management for traction drives in combination with tractor to attachments, e. g. agricultural implements. implement driven process optimization and state-of- the-art sensing technologies. Traction drive systems, e. g. on construction machines, passenger busses, etc.. Relative to tractors, the ongoing developments will show significantly increased power densities based The future availability of electric energy storage on lightweight chassis and high output drivetrains. systems, has been labelled as a separate fourth area Relative to implements, improved operating of application for electric drives. The availability of strategies and new sensing technologies discovered energy storage systems is going to trigger innovative are enabling significant productivity improvements. In solutions, from complementing conventional vehicle combination this is getting tractor/implement systems configuration thru hybrid structures to completely new

92 ICPC 2019 – 3.4 closer to the performance of self-propelled (Isolated Gate Bipolar Transistor), efficient and cost- machinery. attractive power transistors, in the late 1980s.Available since the late 1980s IGBTs Particularly distributed traction is approaching the represent the back bone of inverters. Broad efficiency of self-propelled machinery while application in industrial automation was followed by maintaining versatility and year-round usage of the automotive volume products from the early 2000s. investment. In the meantime, electrification technology has been ELECTRICAL POWER FOR adapted into off-road equipment. Agricultural tractors AGRICULTURAL IMPLEMENTS with higher power electric drive systems were introduced by John Deere with 7430/7530 E-Premium (2007) [2], Belarus’s model 3023 (2009) and Fendt with the X-Concept tractor (2013).

ELECTRIC DRIVE SYSTEM ARCHITECTURE

In current mobile applications, the common source to power an electric generator is a combustion engine. The generator is either mechanically linked to the crankshaft directly or via a transmission (see Figure 2)

In order to compensate for changes in rotational Farmall 450 with IH ElectrAll, 1954 speed and therefore in output voltage and frequency of the generator, a first inverter is applied. Out of this While the first approach has been initiated already DC-link, a second inverter, dedicated to the load, is about 60 years ago, in 1954 when IH introduced the applied to power the electric motor. This motor is IH Farmall 450 with an integrated electric power driving a load either directly or via a transmission or generator (see Figure 1), the IH ElectrAll system, gear set. The architecture is well established and mobile electric drive applications became significantly introduced in various on- and off-road applications, more attractive with the introduction of IGBTs e.g. [4]..

Electric drive system, schematic

VOLTAGE LEVELS, INTERFACES AND based data interface and a 60amps capable power STANDARDIZATION interface are applied.

If power needs to be provided to agricultural The overall power available for implements out of the implements on a 12V level, ISO 1724 connectors, the legacy 12V system is principally limited by the standard interface for lights and the ISOBUS capabilities of alternators and the electric current Connector (see ISO 11783), a combination of a CAN- required. Since the electric current drives the

93 ICPC 2019 – 3.4 conductor size and therefore weight, space and costs efficiently deliver draft at 10 km/h field speed of an electric drive system. represents a benchmark.

In general, for mobile machinery applications, two With the new technology, electric traction assist on voltage classes are considered: implement axles, transferring about 100HP electrical power from the tractor to an implement and using the Class A <50VAC; <75VDC, covering 12/24V implement’s weight for the tractive force allows systems as well as the new 48 V scaling down the tractor’s weight to 20.2 tons. The systems result is a combination of reduced soil pressure, and increased payloads (based on system gross weight Class B 50VAC…1000VAC; 75VDC…1500VDC regulations). An additional opportunity comes via the electrical power bypassing the tractors drivetrain on Class B voltage covers the mainstream of electric and its path from the engine-driven generator to the hybrid drive systems currently present in the implement’s traction assist axle – the provision of marketplace for vehicles and mobile machinery. additional engine boost power becomes possible [4]. Respective requirements are basically covered by the “low voltage directive” (2006/95 EC). Since the term Based on complex simulation-based research in “Low voltage”, defined relative to cross country power 2015, the value coming with electric traction assist lines, is easily misunderstood, the automotive has initially been validated in the field with a industry is commonly using the term HV “Hochvolt” or combination of a mid-size tractor and a slurry tanker. “Higher Voltage” to characterize systems beyond the The power has been generated by a tractor front-PTO common 12V/24V onboard systems. driven module [4].

Geometry and performance of an interface to transfer Complementary field validation has been carried out class B power from tractors to implements has been in the subsequent years. In 2018 detailed field described by the AEF (Agricultural Electronics investigations were done based on a current John Foundation). The respective AEF guidelines Deere 8400R and a Joskin slurry tanker equipped represent the foundation of standardization activities with an electric traction assist system. This system currently in process as ISO 23316: Electrical high- contained the key components, the “Traction Kit”: power interface 700VDC/480VAC. • Off-the-shelve truck axle with differential Triggered by first class B applications on tractors • JD Power systems gearbox having entered the marketplace, a safety standard • JD modular high power-density electric motor has been released, ISO 16230: • AEF-HV-interface compatible power harness

New 48 V systems first introduced by the automotive The field experience made transparent that a industry are becoming attractive to off-road significant increase of productivity was made applications due to their cost effectiveness and available thru the capability to pull wider injection component availability. They are very likely going to tools behind a given tractor. Compared to current be an option for intermediate power levels, e.g. 2 … technology systems and using an exemplary farm 20 kW, on board of tractors and on implements. operation as a base for calculations, productivity Safety focused standardization is ongoing, working increases beyond 20 % were demonstrated. The on the draft for ISO 23285. corresponding revenue increase comes with

48 V systems are not in focus of the current paper. • about 15 % reduced machine operating costs, • improved uphill and sidehill driving, TRACTOR/IMPLEMENT TRACTION • optimized guidance performance, DRIVE SYSTEMS and Tractive performance of agricultural tractors is a function of the force between the tire and soil. The • less soil damages. ballasted tractor weight is the major factor contributing to this force. Ballast and weight To support the development of agricultural transferred from implements add additional force to implements with traction assist features, John Deere the tires. Power Systems has introduced the “Traction Kit” at the 2019 BAUMA at Munich, Germany. To A current technology 400HP 4WD tractor (appr. complement the electric machine and the gearbox 295kW) requiring a weight of about 27.6 tons to with power electronics, John Deere Electronic

94 ICPC 2019 – 3.4

Solutions makes an inverter portfolio accessible to STANDARDS the industry. ISO23316 (in process) -Electrical high-power SUMMARY interface 700VDC/480VAC

In the same way mechanical PTO and hydraulics led ISO 16230 -Agricultural machinery and tractors – to increased productivity of agricultural machinery Safety of higher voltage electrical and electronic and particularly tractor/implement systems by components and systems allowing greater flexibility in implement design and introducing complex implements, electrical power ISO 23285 (draft)- Tractors and machinery for systems are leading the way to increased options for agriculture and forestry – 32-75 VDC Systems for Ag process control and automation. Power can be and EMM transferred to remote regions of the implement where there may not be space to route hydraulic lines.

The key building blocks to deliver electric drive systems on Ag machinery, technology, safety measures, components and interface standard are available to the industry.

Implement traction assist systems are a great example to demonstrate the value coming with electric drives’ enhanced control and power distribution capabilities.

“Slower but wider, without tractor weight increase” is leading the way to a new generation of integrated high-performance tractor-implement systems.

Key components needed for electric drive solution are becoming available thru John Deere Power Systems’ traction kit and John Deere Electronic Solutions’ portfolio of power electronics - The Next Generation is at the door.

REFERENCES

[4] [1] Electric drives in mobile agricultural machinery – products and potentials, Dr. Joachim Sobotzik, ICPC 2013, Graz/Austria, 2013 [5] [2] E-Premium: Höhere Spannung in landwirtschaftlichen Nutzfahrzeugen, Dr. Eckhard Buning, Thilo Kempf, Roger Keil, VDI Tagung Landtechnik, Hohenheim, 2008 [6] [3] Dieselelektrisches Antriebssystem in selbstfahrenden Arbeitsmaschinen, T. Herlitzius, W. Aumer, M. Geißler, M. Lindner, 4. Fachtagung Baumaschinentechnik, Energie, Ressourcen, Umwelt, Dresden, 2009 [7] [4] Electrification as Enabler for New Tractor- Implement Solutions. Smart power system for implement traction drives and process drives, Dr. Rainer Gugel, Dr. Barbara Böhm, VDI Tagung Landtechnik, Hannover, 2015

95

ICPC 2019 – 4.1 The initiative of the AG machinery industry for CO2 emission reduction

Dr. Eberhard Nacke, Patrick Ahlbrand Claas KGaAmbH

Copyright © 2019 AVL List GmbH, Claas KGaAmbH and SAE International

INTRODUCTION Over the past two decades, manufacturers of cars, trucks and mobile machinery have devoted a great Various interest groups or political parties may part of their R&D investments into the implementation question the relevance of human influence in respect of exhaust gas regulations for engines with the aim of to climatic change. However, the majority of the a drastic reduction of particulate matter and scientific community is convinced that there are hazardous nitrogen oxide emissions. As a downside means to counteract global warming. Fighting against of this general focus on engines of entire industries, climatic change becomes a responsibility of any many other innovative developments, which do have branch of our economies, notwithstanding that there positive effects on efficiency and emissions, have may be other sectors, which represent a much higher fallen by the wayside. share of climate relevant emissions.. Now, as we concentrate on climatic gas, a It is a fact that a continuously growing world comparable approach to reduce CO2 emissions of population will starve from hunger if agriculture will engines is conceivable. Nevertheless, the result will not increase its productivity by 50 – 100% within the most likely be extremely limited in its impact if we do next 30 years. Still, the agricultural sector is as well not want to risk what we have achieved in terms of responsible as a major source of climatic gas NOx and particle emissions. emissions. Within the agricultural sector, machinery are only Therefore, success depends on the optimal responsible for a small proportion of total emissions. combination of emission reduction and the Notwithstanding, the AG machinery industry is ready preservation and expansion of productivity in to take its share of the responsibility for our climate. agriculture. We have to foster innovations in Despite the fact, that the engine exhaust emits CO2, agriculture to match both objectives. However, there are many more means to reduce CO2 than just innovations do not arise in a regulatory environment, engine optimization, Figure 1.. but rather in a competitive environment.

Figure 1 CO2 efficiency potential

96 ICPC 2019 – 4.1

Holistic Approach temperature, humidity and solar radiation call for a very individual and site-specific optimization of The agricultural mechanization industry is asking for mechanization processes. If a farmer has to follow a holistic approach to reduce emissions. Although static rules and restrictions neglecting the specific climate-damaging emissions result from engine situation on his site, we will end up in a big loss in exhaust, it does not really make sense to look at the efficiency and as well, we will not exploit the full engine and its potential alone. A tractor emits CO2, potential of emission reduction but the very reason is the emission is the implement, which the tractor may be towing behind. The main Machine and machinery chain improvement may potentials lie in all the aggregates of a machine and provide opportunities, but proper realization is mostly in the respective agricultural processes such as dependent on the human factor, being the driver of tillage, sowing, crop protection, fertilization and the machine. Reducing driver’s vulnerability to run a harvesting. The combination of tractor and implement machine ineffectively in respect to emissions is one of is significant, whose harmony promises substantial the major sources to fight climatic gas emissions in savings. Ag mechanization. Even an isolated view at an individual tractor- The EkoTech project implement combination may be misleading, as a reduction in a first process may call for more Under the umbrella of VDMA, a consortium of major emissions in consecutive steps. We have to regard AG machinery companies and related research entire mechanization process chains to reach our institutions are collaborating in the project “EKoTech objectives sustainably. Indeed, agriculture is differing – Efficient fuel use in agricultural technology”, with the from almost all other sectors as we are working in an objective to initiate a movement in the entire industry, environment, which we cannot standardize. Shape of Figure 2. the landscape, differing soils, and varying

Figure 2 Project team

Figure 3 4-pillar approach of the European Manufacturer associations CECE and CEMA

97 ICPC 2019 – 4.1

A movement to take responsibility in a joint fight for often tends to be inefficient - the human factor. fuel efficiency and against climatic change by utilizing Machinery and process designers may find the entire innovative power of Ag engineering instead substantial improvements, but if the operator is not of reducing it to a narrow regulation based well trained or just has more fun using machines compliance to a single threshold. inefficiently, our contribution in the fight against climate change will be minor. The research project, funded by the German Ministry of Agriculture, involves well-known manufacturers Arable production is characterized by different and research facilities. The close collaboration machinery processes throughout the year, being between industry, science and advisory institutions adapted to the specific needs of an individual site ensure a broad and comprehensive view at farmer’s (Figure 4). Reducing emissions could be easy: do options in the future. less. Reducing intensity of work or even omitting some of these process steps will definitely reduce The structure of the research project is based on the CO2 emissions. However, if reducing emissions 4-pillar approach of the European Manufacturer leads to a decline in productivity, we have achieved associations CECE and CEMA: CO2 potential in nothing. The very sense of farming is production, and terms of machine efficiency, process efficiency, any improvement in terms of fuel consumption has to improvement of operator performance, and potential be related to the quantity and quality of food or forage of alternative energy sources (Figure 3). produced. Consequently, it does not make sense to identify CO2 emission progress per machine or Looking at savings potential in entire process chains hectare. The only valid measure can be CO2 instead of concentrating just on engines gives room emission savings per ton of grain or grass produced for improvements in many directions. Respecting the [l/t]. real world in farming or anywhere else, another factor

Figure 4 EKoTech - Agricultural Process Chains

The research project develops the basic methods for quantify the potential of the AG machinery industry in implementing this concept. Region-typical model prevailing crops and typical farming regions. farms are identified and their location and crop rotation-specific process chains (wheat, maize, and There are many opportunities, but we need to grass) mapped in a simulation model. They form the reconcile the need for emission reduction and basis for the investigation of reduction potentials. Due productivity growth. EkoTech shows that promoting to the nature of farming in an environment, which competition is the best way to provide a CO2 efficient cannot be standardized, the simulation approach solution. provides best means to acknowledge reality, but still

98

ICPC 2019 – 4.2 Long haul truck powertrain control for low emission and fuel consumption in real traffic conditions

Alois Danninger AVL List GmbH Copyright © 2019 AVL List GmbH and SAE International

ABSTRACT IMPERIUM project a dynamic eHorizon system is developed and applied in 3 different trucks with a Fuel efficiency and emissions reduction interact with “look ahead” capability of delivering static (topology, each other and vary with the specific vehicle speed limits, curve radii …), learned (truck behavior) application, operating conditions and mission. and dynamic data (speed limit changes, traffic flow, The pre-condition for predictive vehicle control is the construction sites, incidents, local weather…) with knowledge on future velocity profile. Within the respect to the road ahead [2]. IMPERIUM [1] project a dynamic eHorizon system is developed and applied in 3 different trucks with a The measures for fuel consumption reduction are “look ahead” capability of delivering static, learned grouped into 4 clusters with each combining and dynamic data with respect to the road ahead. The conventional, rule based and/or predictive measures for fuel consumption reduction combine implementations: conventional, rule based and/or predictive implementations. These are engine and EAS control, • Engine and EAS control hybridization, thermal management and global • Hybridization powertrain and vehicle supervisor. A mission- and • Thermal management model- based validation of improvements based on • Global powertrain and vehicle supervisor simulation including realistic traffic scenarios is implemented. This contribution presents selected outcomes with respect to fuel consumption reduction of the EU- INTRODUCTION funded 3-year research program IMPERIUM and therefore gives a glance on upcoming technologies In the transport sector, the reduction of real driving for CO2 reduction under real driving conditions and emissions and fuel consumption in long haul traffic is attractive from a Total Cost of Ownership perspective one of the main societal challenges. Regulations e.g. in commercial vehicle transport. Euro VI are a baseline with respect to targets in defined test environment conditions, while a clear ENERGY MANAGEMENT SYSTEM need exist to have real driving efficiency measurements to promote the introduction of The task of the energy management supervisor innovative solutions for fuel consumption and controller is to suggest a driving strategy, as well as a emission reductions. control strategy for whole powertrain and its components like combustion engine, electric motor, Fuel efficiency and emissions reduction interact with battery, transmission or auxiliaries. The driving each other and vary with the specific vehicle strategy is to a significant extent derived from application, operating conditions and mission. The information of the road ahead and from information overall objective is the development of new means of about other traffic participants. The main objective of predictive and comprehensive powertrain control in developed driving strategy is reduction of fuel an optimal way, exploiting to the full potential of the consumption and emissions by simultaneously individual systems for each vehicle application and ensuring reasonable journey time and driver mission. acceptance criteria.

The pre-condition for predictive vehicle control is the In theory, this would lead to an extremely complex knowledge on future velocity profile. Within the optimization problem for the whole vehicle including

99 ICPC 2019 – 4.2 component details with many degrees of freedom. consider that the combined solution of the sub Such optimization problems are according to current problems approximates a global optimum reasonably state of the art not possible to solve faster than real- well. Looking at the overall system structure (see time while driving. Figure 7) gives valuable insights in how the formulation of the sub problems may look like. It is One possibility to overcome this issue is to break the shown that all relevant powertrain components, such complex optimization problem into several smaller as engine, Exhaust After-treatment System (EAS), optimization problems (sub problems) for e.g. transmission, battery and electric motor have a direct different system levels, components, etc. These sub connection to the powertrain controller. Additionally, problems can be solved fast with acceptable the velocity optimizer exchanges the predicted load accuracy. This approach relies on the assumption, profile with the powertrain controller. In other words, that the solutions of all sub problems can be realized in the powertrain controller all relevant information in parallel, and that they lead to a solution, which is related to the overall optimization problem is reasonably close to the global optimum. available.

Therefore, the key lies in the identification and As stated above, solving the overall optimization formulation of appropriate smaller problems that fulfill problem directly and in real time is currently not above mentioned assumption. The formulation must possible with known state of the art methods.

System Structure

The powertrain controller coordinates all information Based on the collection and evaluation of all optimal exchange between powertrain components and the and alternative operation strategies that the velocity optimizer. Exchanged information contains components and the velocity optimizer provide, the predictions about the future vehicle states (e.g. best overall strategy is selected based on a expected load, expected gear, ...). The predictions parameterizable cost function. Information exchange are tailored to the needs of each component. between the components is updated according to the Components can use the predictions to compute selected strategy. optimal operation strategies for themselves. The overall optimization problem can be solved It is important to note, that if each component only based on the solutions of several sub problems on computes the best operation strategy for itself, these component level. strategies may be conflicting (e.g. EAS needs a higher engine load to increase exhaust gas The overall costs are specified by the cost function: temperature, but the velocity optimizer needs to slow down because of a preceding vehicle). Therefore, 푁−1 푁−1 푁−1 푁−1 each component provides alternative strategies in 퐶푔,푣,푒,푎,푡 = ∑ 퐿푔,푘 + ∑ 퐿푣,푘 + ∑ 퐿푒,푘 + ∑ 퐿푎,푘 addition. The powertrain controller coordinates the 푘=0 푘=0 푘=0 푘=0 possible “search range” of all components to reduce 푁−1 the number of contradicting strategies. An “efficiency + ∑ 퐿푡,푘 factor” or “cost” is assigned to each strategy, giving 푘=0 an indication about the loss of efficiency for that component, if an alternative strategy is used. 푘 corresponds to discrete points in time, 푘 휖 푁+.

100 ICPC 2019 – 4.2

푁 refers to the length of the prediction horizon, exceed the normal target velocity within legal limits 푁 휖 푁+. and consequently increase its kinetic energy just by reducing the braking forces, see Figure 8. The

퐿푔,푘 refers to the cost of the gearbox (friction losses increase in kinetic energy will be used in the and potential loss of travel time), at time 푘 if the gth subsequent uphill part, and therefore the required strategy is applied; 푔 = 1, 2, … 퐺, whereas G is the engine power as well as the fuel consumption can be number of investigated strategies in for the sub- reduced. Thus, a key aspect for fuel consumption optimization problems of the gearbox / transmission. reduction is the computation of an optimal vehicle velocity profile. This is done by the velocity optimizer.

퐿푣,푘 refers to the cost of the velocity optimizer (energy required on vehicle level), at time 푘 if the vth strategy is applied; 푣 = 1, 2, … 푉, whereas V is the number of investigated strategies in the sub-optimization problems of the velocity optimizer.

퐿 refers to the cost of the engine (fuel consumption, 푒,푘 Velocity Optimizer emissions), at time 푘 if the eth strategy is applied; 푒 = 1, 2, … 퐸, whereas E is the number of investigated The velocity optimizer suggests an optimized velocity strategies in for the sub-optimization problems of the profile based on available information within the so- engine. called eHorizon of the road ahead. Based on this velocity and power profile, other optimizers e.g. the 퐿푎,푘 refers to the cost of the EAS (e.g. AdBlue powertrain optimizer compute their optimal strategies. Consumption), at time 푘 if the a-th strategy is applied; 푎 = 1, 2, … 퐴, whereas A is the number of investigated Target of the velocity optimization is the reduction of strategies in for the sub-optimization problems of the the overall required energy, which can be measured EAS. in total fuel consumption (e.g. Diesel & AdBlue) and battery state of charge. Also, aspects like thermal 퐿푡,푘 refers to the cost of the thermal system (e.g. energy management or EAS will be considered. Due increased aerodynamic losses if grill shutter is open, to time constants for thermal measures, which are in or power consumption of pumps), at time 푘 if the tth the range of approx. 10 minutes, a velocity profile for strategy is applied; 푡 = 1, 2, … 푇, whereas T is the this upcoming time interval is required. number of investigated strategies in for the sub- optimization problems of the thermal system. The information of the road ahead is available within the eHorizon. Thus, the velocity profile can rely on 퐶푔,푣,푒,푎,푡 is the overall cost as a combination of the quantities like, e.g.: individual operation strategies as determined by the individual subsystems / components. • Current position or velocity, • Road type (e.g. class, number of lanes, Note that as explained above some of the bridge/tunnel), combinations of individual strategies in 퐶푔,푣,푒,푎,푡 may • Road information (e.g. curvature, crossroads), be contradicting each other and therefore must be • Altitude and inclination, omitted. However, as the powertrain controller takes • (Legal) Speed limits, care to coordinate the potential range of individual • Traffic information (e.g. traffic density or average solutions. The remaining, implementable speed of traffic, traffic jams). combinations are summarized in 푐푔̃ ,푣,푒,푎,푡. The cost of the optimal strategy 퐶푚푖푛 and the corresponding Consequently, based on legal speed limits, average indices of the individual sub strategies on component traffic velocity, curvature and other information, an level 푔, 푣, 푒, 푎 and 푡 can then be determined by effective speed limit can be estimated, i.e. a maximum realizable velocity by the vehicle, which is 퐶푚푖푛 = 퐦퐢퐧 푐푔̃ ,푣,푒,푎,푡 one boundary condition in the optimization. 품,풗,풆,풂,풕 In real life the optimal velocity with respect to fuel VELOCITY OPTIMIZER consumption would be at a very low velocity. The faster a vehicle is driving, the higher is the fuel The velocity optimizer ensures the best use of kinetic consumption due to increased driving resistance and energy. For example, if a vehicle drives downhill, and air drag. Thus, additional boundary conditions with if it is known, that another uphill part will follow respect to a minimum required velocity as well as to immediately, the vehicle could increase the velocity, journey time must be considered.

101 ICPC 2019 – 4.2

Solving the optimization problem on the vehicle while Eco-roll is running on standby while predictive cruise driving requires an optimizer, which is much faster control is active and constantly looks for specific than real time. A two-step approach is implemented events. Such an event may be the detection of a to overcome this issue. At first, based on a vehicle preceding vehicle, which requires a specific action model which contains simplified engine and (e.g. reduce the velocity in a specific way to follow the transmission control strategies, the optimal velocity preceding vehicle at the same speed at a desired profile is computed for a catalogue of different distance). Another example for an event may be the representative driving maneuvers combined with detection of an upcoming curve, for which the velocity altitudes. Secondly, based on these simulations and also must be reduced in a specific way. Eco-roll in gained insights, another real-time capable control is general is not limited to situations where the velocity derived, which results in a close to optimal velocity at the end is lower than at the beginning, though such profile for any arbitrary driving scenario. The real-time cases may occur more frequent. capable control is implemented in the “velocity optimizer” onboard. If an event is detected by Eco-roll, targets for the velocity, covered distance and travel time are Velocity change requests can come from other computed. For the example of approaching a curve, optimizers, which use the velocity profile as basis, that means that the velocity at the beginning of the (e.g. in a thermal optimization loop a higher velocity curve is defined by the “safe cornering speed”, the could be more convenient) or from adaptive cruise distance is specified by the position of the curve and control, for example if another slower vehicle is the travel time is constrained by a driver’s acceptance detected by the radar, see Figure 9. The velocity criteria (drivability) and deceleration limits. A real-time change request needs to be combined with an capable algorithm then finds the optimal (in terms of efficiency factor or cost. The velocity optimizer can fuel economy and travel time) sequence of modes now estimate what following the velocity change tailored to the current driving situation. Figure 10 request would cost in terms of fuel consumption, and shows an example of a scenario for which Eco-roll is it can update its velocity profile based on this value as used. well as on the priority.

Velocity Optimizer connection to adaptive cruise control. Exemplary scenario for Eco-roll. The predictions and suggested velocity profile are made available to the Powertrain Controller and all PREDICTIVE GEARSHIFT other subsystems or components (e.g. engine, transmission control) that may use this information to The Predictive Gearshift Module (PGS) provides gear make predictions and optimizations of their own. The shift suggestions based on the eHorizon road profile velocity optimizer provides the following information: and the optimized vehicle speed from the velocity optimizer. A major aim of predictive gearshift • Predicted vehicle velocity trajectories, (amongst other such as improvement of fuel • Selected vehicle velocity trajectory, consumption) is avoiding drops of the vehicle velocity • Current velocity request, on uphill sections. Therefore, the up- and downshift • Eco-roll request (e.g. if neutral gear is requested) commands should comply with hysteresis and dynamic requirements for driving on flat road, uphill ECO-ROLL and downhill as well as engine speed limits.

Eco-roll is a measure to balance mission time with The PGS module receives information from the energy consumption. It contains an online eHorizon (e.g. road type, altitude, inclination, speed optimization method that automatically finds optimal limits, traffic information, …) as well as an optimized drive mode sequences for specific situations, which velocity profile from the velocity optimizer. predictive cruise control may not handle on its own. In addition to the functionality for the basic shifting shifts scenarios, which are relevant for the powertrain strategy, the PGS calculates costs for possible gear controller.

102 ICPC 2019 – 4.2

These costs consider following aspects: horizon). Further on the desired tailpipe NOX mass flow, desired DOC upstream temperature and an EAS • Gear box efficiency heat up request are provided. • Comfort • Number of gearshifts (especially during uphill) Optimization problem • Foresight shifting The main trade-off parameters for the engine / EAS • Deviations from the received target vehicle speed system which need to be considered and optimized • Travel time (based on driver requests) towards lowest operation costs which keeping compliant with emission limits are: Gear shifts during uphill driving may lead to very high costs, whereas shifting on flat road or on slight slopes • NOX / Total Fuel Consumption (Diesel & AdBlue) may increase the costs only marginally. trade off; • EAS thermal management: Tailpipe NOX versus Possible shifting scenarios/modes of the PGS module engine out temperature as main influence on are: SCR efficiency; • Cooling actuator power versus engine mainly • Efficiency (max. gearbox efficiency) influenced by intake manifold temperature and • Comfort (acceleration and deceleration as low as EGR temperature; possible) • Thermal management / engine coolant circuit: • Balance (mix of Efficiency and Comfort) Power/Temperature engine coolant circuit versus • Dynamic (lowest possible gear, may be also ICE fuel consumption & BSNOX required for predictive thermal control) • EcoRoll-Light (highest possible gear, may be also NOX / Soot tradeoff is considered as not critical for the required for predictive thermal control) considered application. • EcoRoll (Neutral Gear – coasting mode) Cost Function design for optimized predictive The PGS modules calculate for each scenario a engine operation vector with decided gearshift positions and related For the engine cost function primarily, the costs. Therefore, predicted gears for at least two consumption of Diesel and AdBlue is considered. scenarios and the related cost is provided back to the powertrain controller. Additionally, a too low exhaust temperature will be penalized as a constraint, whereas only a negative MODEL PREDICTIVE ENGINE & deviation to a minimum temperature threshold will EXHAUST AFTERTREATMENT affect the cost function. The temperature threshold CONTROL depends on the SCR efficiency. Furthermore, the NOX emissions are considered also The main task of the Model Predictive Engine Control as a constraint, a deviation to a maximum allowed within the energy manager is to provide cost factors limit value will be penalized. for a set of possible engine load points (engine torque/speed). This information is used for the The modelled consumption of Diesel and AdBlue are selection of the gear level and or powertrain strategy accumulated over the prediction horizon to derive the (e.g. E-Motor / engine torque split or load point shift) consumed masses, which then are multiplied with the by the Powertrain Supervisor. The optimized request price per mass. Furthermore, other relevant system values contain the EAS heat up request, the states are modelled/predicted ahead, like engine out optimized request value for tailpipe NOX mass flow temperature and temperatures along the EAS path, and the optimized desired DOC upstream which are then considered in the cost function J. temperature. The optimization is done iteratively, calculating / Additionally, the gear information, torques & speeds modelling the relevant states for each considered are received from the Predictive Gearshift Module. variation along the prediction horizon. For at least two predicted possible torque/speed pairs the corresponding engine and EAS optimizer Following input parameters are varied and the calculates the corresponding cost functions. corresponding cost functions are calculated:

The main outputs are the engine / EAS system total • Engine speed / torque pairs (load points) costs [g/kWh] or normalized cost factor and the trajectories for the preselected gear level velocity offset request to velocity optimizer and EAS sequences in a pre-optimization loop heat up request (time based vector over prediction

103 ICPC 2019 – 4.2

nd • EAS temperature and Engine Out NOX in a 2 The VECTO Long-Haul CO2 assessment cycle is optimization loop used as a basis for assessment. The specific cycle definition used here is from VECTO 3.3.0.125 [3]. The The boundary condition of exhaust gas temperature loading condition used was the Long-Haul Reference to be as low as possible, but as high as necessary to Load (RL-LH). This load condition assumes a ensure SCR efficiency needs to be respected. An 19300kg payload and 7500kg trailer in addition to the online tailpipe NOX observer ensures compliance with tractor mass. The VECTO Long Haul cycle is defined real life emission limits e.g. the In Service Conformity as four vectors of distance, gradient, velocity target regulation. and stop duration.

퐽푀푖푛 = 푚푖푛{C푑푖푒푠푒푙 + C푎푑퐵푙푢푒 + max(0, NOx − NOxLim) Traffic is added to the simulation environment by − (min(TScrLim, TScr) − TScrLim) means of traffic micro-simulation, which employs agent-based modelling to produce naturalistic traffic ∙ PTDevScr} flow patterns. Each vehicle in the simulation obeys a set of rules that govern the vehicle and driver’s 퐽푀푖푛 Overall engine / EAS cost function behavior with respect to the other vehicles in the C푑푖푒푠푒푙 Cost Diesel (accumulated mass Diesel simulation. As each vehicle interacts with the multiplied by price per mass) surrounding vehicles, complex traffic flow patterns emerge, such as congestion and traffic jams. The C푎푑퐵푙푢푒 Cost AdBlue (accumulated AdBlue mass multiplied by price per mass) SUMO (Simulation of Urban Mobility) [4] simulator and TraCI4Matlab [5] are used in the results NOxLim NOX Limit, an exceeding of this limit will be presented here. penalized TScrLim SCR temperature limit, temperature level RESULTS below will be penalized by the cost function Currently the measurement initiative is ongoing to PTDevScr Price or penalty factor when SCR validate the results on demonstrator trucks in real temperature falls below limit driving conditions. For this initiative the algorithms are VALIDATION implemented in prototype control units for online testing. The main challenge to judge the fuel consumption reductions achieved is that a significant proportion of To judge the output of the simulation results and the the expected benefit related to technology fuel consumption improvement compared to the improvement will be in real-world traffic conditions, baseline uses the following equation: which can be anticipated due to the electronic horizon [l] infrastructure. It is expected that the vehicle’s FC electronic controllers will be able to handle upcoming Distance[100km] 퐹퐶푟푒푙[%] = ∗ 100 [%] traffic situations more efficiently than a manual driver, 푙 FCRef [ ] who is not aware of upcoming issues on the route until 100푘푚 they enter the line-of-sight. As simulated traffic is included, interactions between the vehicle under test and the traffic will influence the The only option to get reproducible traffic needed for fuel consumption result. Therefore, the statistical the validation is to use a simulation environment. A nature of the results must be considered. further challenge of validation under reproducible traffic conditions is, that even in simulation, the When simulated under traffic conditions on the simulated traffic itself develops an interaction with the IMPERIUM Long Haul cycle, the fuel consumption simulated vehicle under test. reduction incorporating the described technologies is calculated with -5.1 %. Some key points that that were learnt from this activity are: CONCLUSION • each simulation is deterministic and therefore perfectly reproducible, but; The global powertrain optimisation takes care about power, propulsion, thermal system and pollutant emissions. The outcomes are globally optimised set • a small change in the full load curve of the engine values for the powertrain elements like internal or in the vehicle control strategy has a strong combustion engine, transmission, thermal system, influence on the vehicle’s trajectory through the waste heat recovery, battery and e-motor. traffic jam, even if all the other parameters of traffic simulation are kept the same.

104 ICPC 2019 – 4.2

The use of predictive information in the traditional REFERENCES powertrain domain, especially the combustion system enables further improvements. [1] A. Danninger, N. Knopper and E. Armengaud, "Welcome to the IMPERIUM project," 2019. The dependency on traffic directly leads to the [Online]. Available: http://www.imperium- problem of non-reproducible validation project.eu. [Accessed 01 04 2019]. measurements on real roads. Therefore, a simulation [2] A. Danninger, E. Armengaud, G. Milton, J. – based validation strategy is developed. Lützner, B. Hakstege, G. Zurlo, A. Schöni, J. Lindberg and F. Krainer, "IMPERIUM – It can be concluded that fuel consumption increases IMplementation of Powertrain Control for by circa 10% under the studied ‘real traffic’ conditions. Economic and Clean Real driving emIssion and The changes in vehicle control considering upcoming fuel Consumption," in 7th Transport Research road inclinations and the dynamic electronic horizon Arena TRA, Vienna, 2018. data stream lead to reductions in fuel consumption by up to 5.1 %, particularly in case of increased traffic. [3] "Vehicle Energy Consumption calculation TOol - VECTO," European Commission, 2018. [Online]. DEFINITIONS, ACRONYMS, Available: https://ec.europa.eu/clima/policies/transport/vehi ABBREVIATIONS cles/vecto_en. [4] D. Krajzewicz, J. Erdmann, M. Behrisch and L. DOC Diesel Oxygen Catalyst Bieker, "Recent Development and Applications of EAS Exhaust Aftertreatment SUMO - Simulation of Urban MObility," EGR Exhaust Gas Recirculation International Journal On Advances in Systems PGS Predictive Gear Shift and Measurements, vol. 5, pp. 128-138, 2012. [5] A. F. Acosta, J. E. E. Oviedo and J. J. E. Oviedo, "TraCI4Matlab: Enabling the Integration of the SUMO Road Traffic Simulator and Matlab® Through a Software Re-engineering Process," in Modeling Mobility with Open Data: 2nd SUMO Conference, Berlin, Germany, 2014.

ACKNOWLEDGMENTS

The research leading to these results has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement n° 713783 (IMPERIUM) and from the Swiss State Secretariat for Education, Research and Innovation (SERI) under contract n°16.0063 for the Swiss consortium members. The paper represents cooperative results from the consortium consisting of the following partners:

Special thanks go to Ferdinand Krainer and Marc Seljak who intensively worked on this topic and provided the basis for this contribution.

105

ICPC 2019 – 5.1 Impacts of Digitalization on the Ag Industry

Dr. Markus Baldinger, Dr. Martin Follmer Pöttinger Landtechnik GmbH Copyright © 2019 AVL List GmbH, Pöttinger Landtechnik GmbH and SAE International

ABSTRACT and improved sensors and actuators, low cost micro- processors, high bandwidth cellular communication, Digitalization enables major changes and cloud based ICT systems, big data analysis. improvements in agricultural industry. Implements, Agriculture 4.0 stands for the integrated internal and tractors and the farmers office are going to work more external networking of farming operations. This closely together. Various data are produced by each means that information in digital form exists for all machine involved in the process chain and are farm sectors and processes; communication with combined with sensor data from e.g., weather external partners such as suppliers and end stations, satellites and drones. Furthermore, customers is likewise carried out electronically; and sophisticated mathematical models (e.g. based on data transmission, processing and analysis are historical data and forecasts) are used to optimize the (largely) automated. The use of internet-based processes and to support the farmer in decision portals can facilitate the handling of large volumes of making. The combination of all these data and the data, as well as networking within the farm and with enrichment with artificial intelligence offers the external partners [1]. possibility to gain a holistic view on agriculture and to establish new services as well as business models. Agriculture 5.0 is the synonym for the next evolution of farming consisting of unmanned operations and Also several demanding challenges came up with autonomous decision systems. Agriculture 5.0 will be digitalization for both manufacturer as well as end based around robotics and (some form of) artificial customer. Manufactures have to establish new intelligence [1]. competences in the whole organizations and have to find new partners to handle the new technologies. Agriculture is a global business whose primary goal is End customers have to deal with data platforms and to feed 10 billion people in 2050. All of these data security, farm management systems and also technologies make a contribution to the achievement new technologies on their machines. of the goal through using resources more efficiently, being more animal-friendly, enable a sustainable Digitalization offers manifold potentials to optimize production of high-quality food and optimize working the processes, to increase productivity and to make processes [2]. agriculture more sustainable. Also [3] describes the huge potential of digitalization DIGITALIZATION IN AG INDUSTRY in Ag industry and claims the possibility of a tech- driven improvement of crop yields of up to 70% by 2050 using the following key technologies: a fleet of Agriculture 3.0 – Precision Farming – started once smaller autonomous systems to reduce compaction, military GPS-signals were made available for public precision fertilizer, precision irrigation, precision use and offers solutions for guidance (mid-1990s), planting, precision spraying and several other sensing & control (during the 1990s), telematics technologies (e.g., field monitoring, data (early 2000s) and data management (early 80s). The management). The current situation shows 15-20% intention is to give each plant exactly what it needs to yield loss suffered from inadequate fertilizer grow optimally, with the goal to optimize the application and in turn a potential of 50% reduction in agronomic output while reducing the input (‘more with water waste by using modern irrigation systems. less’) [1].

Agriculture 4.0 – Smart Agriculture or synoym also called Digital Farming – started in the early 2010s based on the evolution of several technologies: cheap

106 ICPC 2019 – 5.1

KEY TECHNOLOGIES automation [6]), hardware components (e.g., radar, LIDAR, camera) and software tools & data (e.g., According to [3] industry experts have the image recognition, maps) developed for autonomous understanding that precision fertilizer applications will car technology is going to foster autonomous lead to 18% increase in yields. This can be achieved technologies in Ag industry. This profound basis has by collecting and analyzing several data e.g., weather to be enhanced with agriculture specific use cases data, soil data, drone and satellite data as well as and features [7]. previous yield data in combination with special hardware components (variable rate application). The All the data generated by e.g., drones, satellites, data can be used to generate a map which allows for different sensors and vehicles used in modern farms a subarea specific output quantity of fertilizer. have to be handled with a professional data Precision planting applications enable subarea management or so called Farm Management specific seeding rates which will lead to 13% Information Systems (FMIS). The required office improvement in yields [3]. This can be done using work, field and fleet management as well as several multi-seed planters which allows to combine feed working steps for livestock farming and outdoor properties of different seeds in one planting or work are supported by professional software tools. In variable rate planting based on the specific subarea. combination with mobile devices the farmer is (in real Based on technologies such as image processing time) well informed about all ongoing processes and and highly accurate solenoid nozzles precision working steps and is connected to all relevant data. spraying applications will lead to approximately 4% improvement in yields [3]. This can be achieved by a CHALLENGES OF DIGITALIZATION subarea specific spraying rates. According to [3] precision irrigation is another promising technology An online survey [8] done with more than 2000 which will lead to 10% improvement in yield and offers persons residing in Germany shows that the German the potential to reduce water consumption by 50% [3]. population has increasingly moved away from A very important enabler for the technologies agriculture and thus the level of knowledge about mentioned above is field monitoring using satellite, current agricultural production processes is reversed. drones, weather data and models as well as remote Explaining the potential of the new digital soil sensors. technologies in terms of environmental protection and animal welfare, the majority of respondents At present, automation is also of great importance in responded favorably to modern digital technologies. Ag industry and is being investigated intensively by Furthermore, the majority of respondents indicate an both companies and research institutes. In recent increasing quality of life of farmers. years small autonomous systems which operate in so called field-swarm as well as big and heavy full-scale Another survey [2] done with more than 500 farmers autonomous tractors have been demonstrated. All shows that 88% of respondents believe that digital systems have common intentions: increase precision, technologies can increase resource efficiencies. By effectiveness and sustainablity, become more contrast, 52% believe that digitization itself is the productive and reduce labor costs. Field-swarms offer biggest challenge, and 39% see the associated costs the additional benefit to decrease the negative as the biggest challenge. influences of soil compaction which is responsible for the soil’s ability to hold water, nutrients and air. In the To demonstrate the potential of digitalization in Ag past decades the size of tractors used for agriculture industry so called test fields are going to be has increased continuously, motivated by the established in Germany, Switzerland (Swiss Future increase of effectiveness and the reduction of labor Farm, already started) and Austria (Innovation Farms costs. As a consequence, weight and dimensions of Austria). The test fields will focus on e.g., the modern tractors have reached levels which are close development and testing of technologies on real to prevailing limits of usability as well as legal farms and therefore under real conditions, connecting frameworks. Through the use of a fleet of smaller of various agricultural processes, testing of tractors (field-swarm) experts expect a yield agricultural digital applications (Use Cases), definition improvement of about 13% by reducing the amount of of suitable data standards, assessment of benefits in soil and roots crushed by heavy agriculture economic, environmental and social terms, machinery [3]. Several approaches of field swarms assessment of opportunities and risks in the areas of have been presented in the recent years: very small sustainability, animal welfare and resource units (Fendt Xaver [4]) as well as bigger units with a conservation. Test fields are going to be established standardized 3m working width (Feldschwarm project as a point of contact to demonstrate the new [5]). Much of the efforts done for the development of technologies, to show the benefits, to enable autonomous cars is already transferred to agriculture. knowledge transfer and to make digitalization more Restrictions, tools and methods (such as the levels of tangible [9].

107 ICPC 2019 – 5.1

CONCLUSION

The amount of new technologies that will find their way into agriculture also requires a lot of different (technical) skills. This not only affect the end customers – i.e., farmers or contractors – but also the machine manufacturers and dealers as well as the relevant educational institutions. From the point of view of a implement manufacturer, more cooperation must be entered to be able to handle all these topics. From the development of special sensors to artificial intelligence to evaluate the generated data, to the programming of apps, all of these topics must be competently supervised in future. This will also lead to changes in already established internal organizational processes (e.g., product development process) as well as new organizational units and possibly also new business models.

REFERENCES

[1] CEMA: Digital Farming: what does it really mean?; 2017 [2] Digitalisierung in der Landwirtschaft; BayWa AG; 2018 [3] Precision Farming; The Goldman Sachs Group, Inc.; 2016 [4] Fendt: Wird Xaver die Welt ernähren?; Thiemo Buchner; 2018 [5] The Evolution of Tractor Implement Systems to Modular and Highly Autonomous Machine Systems (field-swarm); Thomas Herlitzius, Jens Krzywinski, Matthias Klingner; 2019 [6] SAE J3016: Levels of driving automation; 2019 [7] From manual driving to full autonomy: An approach to systematically define different levels of automation in agricultural engineering; Norbert Streitberger, Florian Balbach, Eberhard Nacke; 2018 [8] Gesellschaftliche Akzeptanz von Digitalisierung in der Landwirtschaft; Johanna Pfeiffer, Sebastian Schleicher, Andreas Gabriel, Markus Gandorfer; 2019 [9] HBLFA Francisco Josephinum: Innovation Farms Austria; Heinrich Prankl; 2019

108

ICPC 2019 – 5.2 ADAS and Autonomy for Trucks - A look into the future

Ozan Nalcıoğlu, Alper Tekeli, Güvenç Barutçu, Berzah Ozan Ford Otomotiv AS Copyright © 2019 AVL List GmbH, Ford Otomotiv AS and SAE International

ABSTRACT great potential for truck business. The main drivers of higher levels of automated driving are categorized as Each and every year, road transportation demand is safety, efficiency,environmental concerns and increasing in order to meet the requirements of comfort in “Connected Automated Driving Roadmap” developing world conditions. With the increased of ERTRAC [2] which is published in 2019. number of freight vehicles in the traffic, utilization of advanced driver assistance systems enhanced with Ford Otosan focuses on automated driving of heavy- connectivity features bring substantial benefits on commercial vehicles primarily for long-distance traffic safety, transport system efficiency and driver transportation & confined area logistics which will comfort. Ford Trucks are currently equipped with shape the future logistic and transportation eco- several ADAS functions and Connectivity system. The overall roadmap includes not only L1&L2 technologies which strongly supports truck drivers’ ADAS functions which will be implemented in short needs and logistic sector. term, but also Level-3+ functions which will serve for different customer needs to bring efficiency and On top of ADAS and Connectivity Technologies, comfort in near future. Automated Driving on Highways & Confined Areas holds a great potential for truck business which is the Ford Otosan’s automated truck driving roadmap is leading mode of freight transportation. Harmonious consistent with ERTRAC’s recent report considering development of vehicle automation, digitalization and the development paths & implementation timings of infrastructure will pave the way for transfer hub different levels of connected-automated driving model, combining Highly Automated Driving functions functions on commercial vehicles. on “specific approved roads” & “confined areas” with remote fleet and transport management systems. DIFFERENT LEVELS OF DRIVING Ultimately, a new end-to-end logistic ecosystem will AUTOMATION ON TRUCKS be formed with the usage of highly automated trucks and electrified distribution trucks in interurban and More and more advanced driver assistance system urban areas, respectively. (ADAS) functions (L0/L1/L2) – which help with monitoring, warning, accelerating, braking and Ford Otosan is actively involved in several steering tasks- are being implemented on trucks in research&development projects named OptiTruck, order to increase traffic safety and driver comfort, TrustVehicle, PRYSTINE and 5G-MOBIX related with largely driven by regulations and customer interests. Advanced Cruise Control, Tractor-Trailer Combination Parking, Platooning & C-V2X Without a doubt, the recent regulation proposal of technologies which are planned to be introduced on European Commission which is announced at the future products, aligned with the new business end of Q1-2019 will accelerate the implementation of models of the logistic sector. new ADAS functions on trucks.

INTRODUCTION Available ADAS Functions on Trucks Driven by regulations, traffic safety and driver Statistics of EU transportation modals show that road comfort, current trucks are mainly equipped with transportation has the largest share of EU freight L0/L1 functions such as Lane Departure Warning, transport by %75 consistently for the last 5 years. [1] Driver Drowsiness Detection, Advanced Emergency Taking the increasing demand on transportation & current modal splits into account, Automated Driving Braking System, Adaptive Cruise Control, Adjustible enhanced with Connectivity Technologies hold a Speed Limiter Device, Blind Spot Detection. With the new regulation proposal, Driver Drowsiness

109 ICPC 2019 – 5.2

Detection & Blind Spot Detection functions will be • Safer highway; less number of accidents. regulatory as Lane Departure Warning & Advanced • Reduced fuel consumption and air pollution Emergency Braking was in 2015. • Less transportation costs for goods and services

Ford F-Max is equipped with Adaptive Cruise Control Long-haul trucking requires driving day and night function which has extended braking capability on top through varying weather, road and geographical of regulatory functions – Lane Departure Warning & conditions in just one mission. Because of this Advanced Emergency Braking. Thanks to reason, effective automation should include the connectivity technologies and tophographical maps, capability to handle all of these different conditions. F-Max has a Predictive Cruise Control function so According to OEM’s public announcements, state-of- called MaxCruise which estimates the optimal speed the-art commercial solutions for L4 will be available profile using road tophography and drivers’ set speed, by 2020-2021 but will only be able to handle resulting in up to %4 fuel economy as an economy & autonomous driving under certain conditions (low comfort function. speed, day, good weather etc.). That’s why, one of the main focus areas at Ford Otosan is to enable Potential Future Functions on Trucks automated long-haulage in adverse conditions. Technical approach towards this goal is mainly based On top of the above mentioned ADAS functions - with on novel deep learning models that are trained with the upcoming automated driving developments - data collected by trucks under several different Highway Autonomy, Confined Area Autonomy and conditions in Turkey. Ford Otosan’s existing fleet and Connectivity Technologies hold a great business truck usage know-how help feed the AI researchers potential in trucking industry. with good quality data. Together with platooning, predictive cruise control, and other fleet operation Although each three potential areas have their own functionality, L4 autonomy will bring further reduction business and operation models, combination of these of transportation costs and pave the way towards L5 technologies will lead to a new logistic model, so trucks. called hub-to-hub transportation. This new logistic model combines conventional trucks for the first and Although the results achieved in the last decade last mile delivery with automated L4 or even enabled a breakthrough in the world of autonomous driverless (L5) trucks for the middle part on the vehicles, there are still many scientific challenges that highways. Trailer/Load changes will be performed in require extensive research. These challenges may be transfer hubs which has direct access and connection briefly summarized under the below headings: to highways. Confined Area Autonomy takes the stage in transfer hubs while parking, attaching- deattaching trailers and maneuvering, whereas • Intelligence level Highway Autonomy enables autonomous driving and a. Model architecture platooning on highways. To improve overall system b. Training methods efficiency and safety during hub-to-hub transportation c. Data quality model and other business models, connection to the • Adaptability level: Capability of the model/system infrastructure and other vehicles (V2x) is essential to adapt to different conditions (road, weather, and indispensable. geography etc.) • Speed & Processing Resources Required: HIGHWAY AUTONOMY a. Training speed & performance b. Resource requirements for real-time performance Global OEMs are all carrying out various activities in • Safety: Capability of the system to overcome the aspect of “transportation-as-a-service”. It is critical unsafe situations for vehicle manufacturers to grasp the idea behind x- as-a-service and put strategic plans in place. • Security: Vulnerability of the system to Autonomous driving is certainly an enabler of large- cyber/physical attacks scale hub-to-hub transportation, in a way creating • Ethics: Ethical issues regarding training & virtual road trains, and potentially lowering the costs decision making processes road transportation, while increasing the speed, • Transparency: How to track the decisions made safety and reliability. With Level 4 Highway Autonomy during the ride and how to inform the driver / of trucks, following social benefits are expected: passengers about them

• For the drivers, reduced stress levels and ability Ford Otosan is working on creating a holistic to carry out other tasks during the ride. Autonomous Driving SW design & evaluation strategy • Reduced level of traffic congestion; saving time and finding optimal ways to combine the strengths of for all road users. AI and non-AI based approaches. This requires very

110 ICPC 2019 – 5.2 well linked theoretical analyses, simulations and field cumulative result of camera performance, lane testing. marking status, road and weather conditions.

Many of the above aspects need further improvement in order to be able to deliver an L4 Autonomous Long- Haul Truck. With all these unsolved problems, there is an opportunity for the truck OEMS and suppliers, and start-ups to create novelties that enable competitive advantage, in terms of cost, quality or timing. Research at Ford Otosan is mainly geared towards creating competitive advantage in the fields of perception and planning for long-haul trucks under bad weather, road, and environment conditions. Figure 2 Istanbul-Ankara Highway Segment Enabling no-stop long-haulage under adverse Lengths conditions will create commercial benefits for the stakeholders as it also brings safety benefits. Results obtained from 20 Ford F-Max tractors show that lanes can be detected for the %99,52 (average In addition to the technical developments which are of all segments) of the complete highway which mainly concentrated on unsolved problems of L4 means during 2.13 km of the total 445 km, lanes are Autonomous Long-Haul Truck, infrastructural not detectable or camera is not able to detect lanes requirements and customer experience/habits are due to several reasons. Figure 3 and Figure 4 also investigated. For these spesific purposes, more provides some examples for detected and missed than 20 Ford F-Max tractors are observed in a specific lane marking sections of the highway, respectively. route presented in Figure 1.

Figure 1 Istanbul-Ankara Highway

The highway connecting Istanbul to Ankara is selected as the specific route for the above- mentioned investigations since it is the longest and mostly used highway in Turkey according to the information provided in the official website of “Ministry of Transport and Infrastructure”[3]

Considering the regulations/rules being discussed for Platooning function, Istanbul-Ankara highway is divided into 25 segments (from exit-to-exit) to be able to perform an infrastructural analysis more efficiently. Segments have varying lengths from 8.5 km to 47.5 Figure 3 Examples of Good Lane Marking Sect. km with an average of 17.8 km and results in a total on the Highway of 445 km highway. The segment lengths are provided in a bar chart in Figure 2. In addition to the lane detection performances on different segments of the highway, driver Although, Autonomous Driving or specifically workloads/habits are also investigated segment per Platooning on shorter segments does not bring segment to quantify the benefits of potential highway significant benefit on fuel consumption and driver autonomy functions. Accelerator Pedal & Brake Pedal comfort, all defined segments are inspected (or auxiliary brake activation) are frequently used by separately for infrastructural requirements. For each drivers while travelling on the highway and those two segment defined in Figure 2, Lane Detection driving tasks hold a substantial portion in the driver’s Performances are investigated separately which is a

111 ICPC 2019 – 5.2 workload on top of steering and monitoring driving CONFINED ZONE AUTONOMY environment. Highway Autonomy mainly covers autonomous driving between transfer hubs on highways but Autonomous Driving in Confined Areas is complimentary for a complete hub-to-hub transportation, which is the future vision for road transportation & logistics. Considering that tractor- trailer combinations can park at the associated locations in the transfer-hubs autonomously or tractors can attach to the designated trailers autonomously within the transfer hub, the last piece of the hub-to-hub transportation puzzle will be completed.

On top of being complimentary to Highway Autonomy for a complete hub-to-hub transportation; Confined Zone Autonomy has its own potential business models especially in ports, logistics areas and construction/mining areas. Driverless, remotely supervised or even cabless trucks for some specific usages will be possible with Confined Zone Autonomy.

Ford Otosan is currently involved in 4 different EU- Funded projects which have different use-cases Figure 4 Examples of Missed Lane Marking related with Confined Zone Autonomy; TrustVehicle Sections on the Highway [4] & PRYSTINE [5] are concentrated on Auto-Trailer Parking function at Logistic Areas, 5G-Mobix [6] & Figure 5 presents the ratio of accelerator pedal, brake NewControl (will start in June-2019) are taking pedal, auxiliary brake usage to total travel time while Remotely Supervised Trucks on Customs Area & travelling on Istanbul-Ankara highway with ACC-On Construction Area as use-cases, respectively. and ACC-Off conditions, respectively. This information provides an estimate potential of a L2- Auto-Trailer Parking Function @ Logistic Areas Automated Truck Platooning and L4-Highway Pilot Platooning on driver workload reduction. One of the most challenging driver tasks on tractor- trailer combinations is the reverse parking or so called backing to a docking station in a logistic area due to the articulation between tractor&trailer as presented in Figure 6.

Figure 5 Driver Workload Reduction by ACC on Istanbul-Ankara Highway

Even with Adaptive Cruise Control, driver workload Figure 6 Sketch of a Tractor-Trailer Combination generated by acceleration & braking tasks is reduced Backing to a Dock by half which strongly points out that the driver workload improvement will be more with L2-

Automated Truck Platooning on top of its efficiency on fuel economy that the function will bring.

112 ICPC 2019 – 5.2

Ford Otosan is mainly concentrated on the CONNECTIVITY TECHNOLOGIES identification of optimum path for the tractor and trailer during reverse parking and improvement of Connected vehicles will have significant role for tracking performance in TrustVehicle project. The transportation, environment and road safety. Vehicles optimum path calculations are performed based on will communicate with other vehicles (V2V), the robust & fail-safe environment perception infrastructure (V2I) and pedestrian, cyclists or algorithm, so called occupancy grid generated within anything (V2X). The connected ecosystem will enable PRYSTINE project. In addition to the environmental safer roads by avoiding potential crashes, reduced perception, dock detection algorithms are being congestion by efficient route and speed optimizations, developed in PRYSTINE project as well. Figure 7 lower CO2 emissions by less fuel consumption. presents the initial results of dock detection algorithm. It will be possible to see the beyond the obvious by interacting with the other vehicles, infrastructure such as traffic lights, road side units (RSU) or road users. Instead of limiting the driving capabilities by the on board sensors like radar or camera it will be possible to get wide range of valuable information from the V2X units within the vehicle.

The data could be exchanged between vehicle to everything and the collected and consolidated data would be sent back to vehicle to have better environment understanding. The increasing number Figure 7 Initial Results of Dock Detection of connected vehicles will improve the data accuracy Algorithm and freshness.

Remotely Supervised Trucks @ Customs & Vehicles could get benefit from that real-time Construction Areas connectivity data in various ways. While that information is used to prevent dangerous situations Autonomous Driving of trucks hold a great potential such as traffic accidents any information related to on closed construction areas, mining areas and other weather, road conditions or traffic lights would also be confined areas such as customs. In a strictly used for route or autonomous driving profile confined, challenging and even a dirty environment, optimization. equipping the area with relevant sensors is more efficient, safer, cleaner and cheaper than equipping Truck platooning is one of the use cases which are each and every truck. With this approach the trucks enabled by V2V technology. It is a string of vehicle will be remotely supervised from a control center, which is moving cooperatively by sharing the vehicle which will also enable the optimization of co-operative local information and platoon specific messages task management for each truck or any other through the V2V communication. V2V enables a construction equipment. Figure 8 shows a detailed reduced time gap between the trucks. As a sketch of the remotely supervised trucks at a consequence of reduced time gap, fuel efficiency construction area. increases and CO2 emission reduces.

Ford Otosan and AVL have started a 3-year joint R&D project to develop “Platooning” technology for long- haul trucks. Within that project, for which field tests are ongoing, DSRC based communication is being used as the V2V technology.

In addition to that, Ford Otosan has an active role in 5G Mobix – H2020 project for cross border platooning, platooning see-what-I-see functionality and real time remotely supervised truck routing at a customs site, which will be enabled by 5G based C- V2X technology. Despite the numerous advantages of platooning, from the point of view of the vehicles that are at the back, following the lead vehicle in a Figure 8 Remotely Supervised Trucks at a platoon can cause lack of attention and anxiety while Construction Area driving, since trailers are wide and high enough to

113 ICPC 2019 – 5.2 cover driver sight. To circumvent them, a “see-what- REFERENCES I-see” application will be designed and implemented for truck platooning within the 5G Mobix project, [1] https://ec.europa.eu/eurostat/statistics- which will be providing the road view of the leader explained/index.php/Freight_transport_statistics truck to the others in the platoon. _-_modal_split#Modal_split_in_the_EU [2] ERTRAC-CAD Roadmap-2019 As another use case of the 5G Mobix project, real time remote truck routing will be demonstrated at a [3] http://www.uab.gov.tr/eng/ customs site. One of the key ideas behind remote [4] http://www.trustvehicle.eu/ supervised truck control in this project is controlling [5] https://prystine.eu/ the non-autonomous drive-by-wire trucks by the use [6] https://www.5g-mobix.com/ of external sensors in the field (i.e. customs site, logistics hub etc), instead of the perception sensors on the truck. In this way, the truck will be able to move from one point to another without the driver, DEFINITIONS, ACRONYMS, performing tough maneuvers autonomously in a small ABBREVIATIONS area, which has other vehicles and pedestrian traffic. Sensor fusion in this case will include data from both CAD Connected Automated Driving. the truck and the field. This will enable the driver to save time to expedite other procedures required to get border pass approval while, at the same time, traffic efficiency will be increased, and the average vehicle border crossing time as well as the number of traffic accidents due to human error will be decreased at customs sites.

CONCLUSION

All in all, Automated Driving on Highways and Confined Areas, on top of ADAS and Connectivity Technologies will shape the near future of transportation, logistics and trucking ecosystem. Ford Otosan is developing not only ADAS functions to increase traffic safety and driver comfort with the current technologies but also Automated Driving technologies boh on Highways and Confined Areas to be able to provide an end-to-end solution for potential customers, especially for hub-to-hub transportation and remotely supervised trucks in confined areas.

114

ICPC 2019 – 5.3 Digital Twins – Enabler for Digitalization

Thomas Fischinger, Stefan Hirschenberger, Franz Rimböck Wacker Neuson Beteiligungs GmbH

Copyright © 2019 AVL List GmbH, Wacker Neuson Beteiligungs Gmbh and SAE International

ABSTRACT production models, which are referred to as digital twins. The goal is to convert the physical and virtual At the moment we stand on the edge of a technical world starting from optimized product development, revolution that will fundamentally change the way we processes for new machines to a predictive spare live, work and communicate. The transformation into part delivery for our customers (see Figure 1). At a digital society is taking place at an unprecedented Wacker Neuson we tackle these new challenges that speed. For machinery industry this means to increase rise with the digital transformation by subdividing the flexibility and adaptability of industrial production them in three major streams referred as Smart systems by implementing virtual product- and Processes, Smart Factory and Smart Customer Solutions

Figure 1 Concept of a digital twin as aimed at Wacker Neuson Group

INTRODUCTION lifecycle covers the entire process from product planning, product developing, manufacturing The evolution of the digital twin in industrial processes as well as sales and after-sales service environment has started with the possibilities [4]. computer-aided design (CAD) in the early 70s 1971 by Patrick J. Hanratty [1]. This has revolutionized Along with the developments and applications of new and disrupted the product development as well as the information technologies, a new era of smart production processes. Another corner stone in the manufacturing is coming. By converging the physical evolution has been the concept of product lifecycle world and the virtual world, a series of smart operation proposed by Dean [2] in 1950. This concept was in the manufacturing process becomes possible. This extended to the engineering field by the influence of concept is known as digital twin [5-7]. In this paper the concurrent engineering [3]. Therefore the product

115 ICPC 2019 – 5.3 author describes the concept of a digital twin on design phase the virtual product prototype is created example of Wacker Neuson Group. including product functions and possible product configuration. Essential hereby is an additive-stacked DIGITAL TWIN AT WACKER NEUSON variants configuration. In this phase, our virtual GROUP product design run through several simulation processes e.g. kinematic simulation to optimize our construction equipment. The necessary data is The Wacker Neuson Group develops, produces and obtained by simulation algorithms fed with real-life distributes concrete technology, compaction machine data and defined boundary conditions. The equipment, worksite technology and compact virtual verification of the design substitutes the construction equipment, also offering a range of physical prototype to a large extent. This enables complementary services. Next to the headquarter in accelerated and cost-efficient product development. Munich, the Wacker Neuson Group have seven production and development sites around the world DIGITAL TWIN-DRIVEN, VIRTUAL PRODUCTION employing around 5,500 people. In the next step information coming from product Motivation engineering needs to be translated for manufacturing purposes. At Wacker Neuson Group, we use SAP VE A suitable digital twin strategy enables a company as to bridge the gap between engendering and Wacker Neuson Group to accelerate and de-risk new manufacturing by referencing to the same data product introduction, supports a more flexible and provided by SAP ERP system (see Figure 2). This optimized production process and allows the enables our team to virtually plan the production in implementation of new data-driven services and accordance to the resource allocation such as business opportunities. Unfortunately, a suitable materials, equipment tools, method time software suite is often not available. At Wacker measurements data etc. within the assembly line. Neuson Group we follow a strict concept of After this step purchasing respectly make-or-buy standardization and centralization. The key is to decisions are initiated. Optimization process between establish scalable business processes, that match resource allocation purchasing and logistics is key- SAP standards to the highest possible degree. performance-indicator based. Therefore, we are working with SAP Visual Enterprise (SAP VE) as a backbone for the digital twin in DIGITAL TWIN-DRIVEN PRODUCTION AND production and add functionality with the help of self- ASSEMBLY designed Fiori-apps. This enables us to adapt in The production and assembly process starts with a accordance to the specific needs of our historically model sequence planning developed in-house to fit grown production sites. the requirements needed (see Figure 2). This tool automatically generates an optimizes an assembly Digital twin-driven processes sequence in accordance to historical data, assembly DIGITAL TWIN-DRIVEN, VIRTUAL PRODUCT rules and customer orders. The algorithm is realized DESIGN within a SAP Fiori app to ensure a smooth interface Traditional product design requires professional to our SAP ERP system and a maximum of flexibility. knowledge and expertise of each part of the process. After starting a new machine within an assembly line, In recent years, the processes in product this machine is been tracked by a geo-tracking development have become more virtualized. Starting system based on ultra wide band technology. So the in the conceptional design phase, the designer define progress of the assembly is documented at anytime the concept of the new product. The input of the digital and linked to our quality control systems, which is also twin and data regarding customer satisfaction, based on pure SAP standard functionality (see product usage, product sales as well as other Figure 2). information supports the designer. During the detailed

116 ICPC 2019 – 5.3

Figure 2 Extract from the systems involved

To provide employees with a shop floor suitable DIGITAL TWIN-DRIVEN FIELD USAGE AND frontend we count on custom developed Fiori Apps SERVICES once more. Wacker Neuson Group use variant In order to be able to establish itself permanently on specific checklists to ensure product quality within the the market in a rapidly changing market and in times manufacturing line. Serial numbers of function critical of globalization, it is important to know in detail its parts get recorded and possible problems within the market, its customer segments, as well as the assembly line are documented . The data which is customers themselves and their requirements. In our gathered throughout the whole assembly line developments, we rely on the involvement of our enriches the digital twins of the single product customers at an early stage to get a voice of the instances with product lifecycle management (PLM) customer. The different systems and possibilities of relevant information. Due to the progress tracking in digitization and the internet of things enable us to the line, the pre-assembly can be triggered precisely. develop our products and services even more Due to the real-time changes in the assambly line, the efficiently and in a more target way. system need an iterative adjustment and optimization. At the heart of this process are the integrated telematics modules and the corresponding app The possibility of incorrect assembly is prevented by landscape. The development of a solution for a cross- an automatic plausibility check, which uses data from product fleet management system for light and SAP ERP. Another part of the digital twin in compact construction equipment is attributed to the production is the worker-guidance system (WGS), end product. For Wacker Neuson Group it is which provide the assambly worker with necessary important to offer a total solution for all equipment information. Next to general information regarding the involved in the construction process. Next to the machine, the worker gets specific data about options, medium and large equipment that can be connected safety instructions, quality instructions as well as to the CAN-BUS system, the light construction engineering change notices. Due to the possibility of equipment has its own challenges. For example, a providing feedback to these topics within the WGS, gasoline-powered rammer has no power supply of its the digital data-set regarding the product or own. There are also special requirements due to the production design is enriched. high vibration. In these applications, a possible technical solutions can be realized via a bluetooth low

117 ICPC 2019 – 5.3 energy beacons in conjunction with a corresponding REFERENCES app on mobile devices. [1] Hanratty, P. J. (2005) MSC: Company History. Using different techniques, the all machines Retrieved from automatically login on their own in case of https://web.archive.org/web/20050209155207/ht maintenance, low levels, malfunctions, unexpected tp://mcsaz.com/about/founder.htm relocation, etc. – in real-time on mobile phones or [2] Dean J (1950) Pricing policies for new products. desktop PCs. This enables our customer to minimize Harv Bus Rev 28(6):45–53 machine downtime, optimize repairs and maximize the return on the machine. [3] Nevins JL, Whitney DE (1989) Concurrent design of products and processes: a strategy for The backend of our fleet management and service the next generation in manufacturing. McGraw- management oriented products is the Wacker Hill Companies, New York Neuson Cloud. Within this system, the data are [4] Ryan C, Riggs WE (1996) Redefining the collected and processed for the customer. In addition, product life cycle: the five-element product Wacker Neuson Group is able to offer new business wave. Business Horizons 39(5):33–40 services such as pay per use, solution as a service, [5] Glaessgen E, Stargel D (2012) The digital twin etc.. The machine data prove a good insight into how paradigm for future NASA and US Air Force the machines are used in field and how Wacker vehicles. 53rd AIAA/ASME/ASCE/ AHS/ASC Neuson can improve the product development to fit Structures, Structural Dynamics and Materials these requirements (see Figure 1). Next to the Conference 20th AIAA/ASME/AHS Adaptive improvement of our products, the data reveal Structures Conference 14th AIAA 1818 information about future behavior. [6] Tao, F., Cheng, J., Qi, Q. et al. (2018) Digital twin-driven product design, manufacturing and CONCLUSION service with big data. Int J Adv Manuf Technol 94: 3563. With the opportunities offered by a world of ever faster [7] Q. Qi and F. Tao (2018) Digital Twin and Big technologies, systems and processes must also be Data Towards Smart Manufacturing and Industry adapted at the same pace. Therefore, the digital twin 4.0: 360 Degree Comparison. IEEE Access, vol. of a process will always be a virtual picture 6, pp. 3585-3593 corresponding to the options available. In this paper the author explained the current state of the attemps to digitalize our PLM data within a consistant system landscape at the Wacker Neuson Group. In the next DEFINITIONS, ACRONYMS, steps the data will be used for smart business ABBREVIATIONS services and analytical services (see Figure 3). CAD Computer-aided design SAP VE SAP Visual Enterprise WGS Worker Guidance System PLM Product lifecycle management

Figure 3 Overview of the main system components of the digital twin at Wacker Neuson Group

118

ICPC 2019 – 6.1 Powertrain Trends and Developments in the Commercial Vehicle Industry

Dr. Carl Hergart PACCAR Technical Center

Copyright © 2019 AVL List GmbH, Paccar Inc. and SAE International

ABSTRACT occurred in the area of system integration. Today’s powertrains are the result of a sophisticated interplay The last several decades have seen dramatic between the major components: engine, improvements in the performance, fuel efficiency and aftertreatment, transmission and axles. Advanced environmental footprint of powertrains for commercial control algorithms run as a thread through the applications. Today’s diesel engines are more power operation and actuation of the subsystems. dense, efficient and clean than they have ever been. Powertrain components are carefully selected with Yet, the challenges for the future are significant. With the overall mission of the truck in mind. Continued increasingly stringent requirements up to and pressure to improve transport efficiency and reduce including zero emission operation, powertrain the environmental impact of commercial powertrains development continues unabated. will drive future developments. The upcoming North American Greenhouse Gas Phase 2 regulations call Substantial advances in the efficiency of engines, for reductions of up to 27% (Figure 1), much of which transmissions and axles notwithstanding, perhaps the will come from improvements to the powertrain. The most notable powertrain development over the last 10 subsequent sections of this paper will discuss some years has been the seamless integration of all of the key technology improvements and future components. This paper will review past, current and opportunities for the major powertrain components future developments aimed at evolving not only and subsystems. conventional diesel powertrains, but also those incorporating increasing levels of electrification.

INTRODUCTION

The importance of commercial trucking cannot be overstated. Over 10 billion tons of freight is transported on trucks every year in the United States, representing 70% of domestic tonnage. 7.7 million people were employed in jobs relating to trucking activity in 2017 in the U.S. [1]. Applications span a spectrum of vocations and it is important to understand the different use cases. Similar to the automotive sector, the commercial industry is very much in the spotlight with regards to emissions Figure 1 North American Greenhouse Gas Phase reductions. Regulatory and competitive pressures 2 regulations for two tractor have driven trucks to become increasingly clean and configurations1 efficient. Improvements to the individual powertrain components and subsystems have been substantial, but perhaps the most remarkable developments have

1 C8 SC HR = Class 8, Sleeper Cab, High-Roof; C7 DC MR = Class 7, Day Cab, Mid-Roof

119 ICPC 2019 – 6.1

ENGINE TECHNOLOGIES engine back pressure. Included in the figure is also the 55% SuperTruck II target established by the U.S. The most direct way of improving the efficiency of Department of Energy. diesel engines is to increase the mechanical yield from the chemical energy stored in the fuel. Hence, As BTE inches up towards the Carnot theoretical limit, there is active research ongoing to optimize the researchers also turn to ways in which to harness the combustion system. The design of today’s energy contained in the exhaust. Such Waste Heat combustion chambers is the result of a century worth Recovery (WHR) methods include turbo- of practical experience enhanced by sophisticated compounding and the Organic Rankine Cycle (ORC). high-fidelity simulations, optimizing such parameters as piston bowl shape, injector geometry and port geometry. Improvements in material properties have enabled increases in peak cylinder pressure and turbocharger speeds, ultimately increasing efficiency while complying with ever more stringent emissions standards. Figure 2 shows the evolution of engine tailpipe Nitrogen Oxides (NOx) and Particulate Matter (PM) standards for Heavy-Duty commercial vehicles in North America over the past 35 years, along with some of the key enabling technologies. From 1985 until 2002 Heavy-Duty onroad emissions standards in North America were largely achieved by optimizing and calibrating in-cylinder performance parameters. Figure 3 Progression of peak engine Brake In 2002-2004 Exhaust Gas Recirculation (EGR) was Thermal Efficiency in HD trucks over the introduced and 2007 saw the introduction of exhaust last 60 years gas aftertreatment systems. As part of the SuperTruck II program, PACCAR is exploring technologies such as the Miller cycle, low friction materials, thermal barrier coatings and ORC to achieve an unprecedented 55% BTE [2].

Modern diesel engines are equipped with an aftertreatment system, consisting of a Diesel Oxidation Catalyst (DOC), Diesel Particulate Filter (DPF), Selective Catalytic Reduction (SCR) catalyst and an Ammonia Slip Catalyst (ASC). The system has to be designed in such a way as to optimize the overall system efficiency while meeting prevailing emissions standards. Figure 4 shows the familiar trade-off between BTE and engine-out NOx (solid

lines). The dotted set of lines take the total fluid Figure 2 Evolution of Heavy-Duty onroad NOx and consumption into account, illustrating that an PM emissions in North America (source: optimum can be found for engine-out NOx. DieselNet) Interestingly, this optimum occurs at different engine- out NOx-levels depending on the cost of Diesel At the same time efficiency has improved. Figure 3 Exhaust Fluid (DEF)2. The trade-off in Figure 4 also shows how the peak Brake Thermal Efficiency (BTE) illustrates the challenge of continuing to improve of diesel engines in commercial trucks has evolved engine efficiency in the face of regulatory reductions 3 since 1960. We note that efficiency has continued to in NOx , potentially requiring reduced engine-out NOx. increase while engines have become much cleaner. The dip in efficiency around 2002 is attributed to the introduction of EGR, which required increasing the

2 It is interesting to note that the ratio of DEF to Fuel is 3 Per a recent staff white paper, the Low NOx regulation is likely to approximately 65% in the U.S. vs. 30% in Europe, making it more be set in the range of 0.05-0.08 g/bhp-hr. CARB is also proposing attractive to pursue higher engine-out NOx in Europe than in the US to implement a new Low Load Cycle over which compliance has to from a total fluid consumption perspective. be demonstrated.

120 ICPC 2019 – 6.1

controls and actuation have become more sophisticated and driver demographics are changing, recent years have seen a noticeable increase in the use of Automated Manual Transmissions (AMTs). Unlike automatic transmissions, AMTs do not have a torque converter, which make them the preferred choice for efficiency conscious line-haul customers. Automatics are finding more widespread use in vocational applications, where the powershift capability is an important attribute. Automated shifting requires seamless integration between engine, clutch, and transmission. Shift strategies are developed allowing the engine to operate in its most Figure 4 Fuel/Fluid Consumption vs. Engine-Out efficient region, while having sufficient power to climb NOx hills without resorting to excessive shifting or gear hunting. Features such as launch and low-speed Downspeeding has long been discussed as a means maneuverability require a delicate interaction to improving efficiency. For transmissions and axles between engine, clutch and transmission. The there is a pretty straightforward relationship between recently introduced PACCAR Transmission (Figure 5) efficiency and downspeeding, whereas it is a bit more is representative of a new class of Automated complicated for the engine. In a downsped engine, Transmissions, which - unlike traditional AMTs – were the combustion takes place within a shorter crank never intended to be manual and as such are not angle window, which gets the overall process closer constrained by a shift pattern conducive to manual the ideal thermodynamic constant volume cycle. The operation. This creates additional degrees of turbocharger also tends to operate more efficiently at freedom, allowing a lightweight, compact and highly reduced speeds. On the other hand, in order to efficient design. maintain power output, a downsped engine has to develop higher torque, which is typically accompanied by increased cylinder pressures and thermal loads. The core engine design and overall powertrain configuration become very important in the ability to take advantage of downspeeding.

Future technology improvements, such as advanced combustion concepts and low friction materials will potentially change the engine speed at which peak efficiency is achieved.

TRANSMISSION TECHNOLOGIES

The basic role of the transmission is to transfer power from the engine to the drivetrain, such that overall powertrain efficiency is maximized. There are some key factors to consider when it comes to selecting the right transmission for a given application: Figure 5 PACCAR 12-speed Automated Transmission • Overall gear ratio • Transmission type Relative to the question of whether to spec a Direct • Direct Drive vs. Over Drive Drive (DD) or an Over Drive (OD) transmission, the former has historically been the preferred choice for The overall gear ratio should be chosen to achieve Over-The-Road (OTR) applications primarily high efficiency at cruise conditions, while ensuring exposed to lower speeds (≤ 62 mph), moderate loads adequate startability and gradeability. Furthermore, (< 80,000 GCVW) and flat terrain. Under these the steps between the gears may be progressive in conditions, having a top gear direct connection order to keep variations in engine speed to a between the engine and the transmission output shaft minimum, especially in the upper gears. results in an efficiency advantage. However, with recent advances in gear design and architectures Historically, the North American market has been involving fewer meshes, the efficiency spread dominated by manual transmissions. However, as between DD and OD has narrowed. As a result, the

121 ICPC 2019 – 6.1 market in North America is increasingly migrating towards OD transmissions.

AXLE TECHNOLOGIES

As with the transmission, the choice of axle is highly application dependent. Some key design choices include:

• Axle configuration • Axle ratios Figure 6 shows a schematic of two different axle configurations: a 6x2 and a 6x4. In North America, Figure 7 PACCAR 40k tandem axle 6x4 dominates OTR applications. A 6x2 configuration has the advantage of lower rolling resistance, but has POWERTRAIN INTEGRATION less traction than the 6x4 and may lead to accelerated tire wear. New technologies are emerging that allow While the individual subsystems have become either 6x2 or 6x4 operation depending on the increasingly efficient, it is the seamless integration of condition. The Dual Range Disconnect™ concept all building blocks that have enabled major strides in offered by Dana [3] allows the tandem axle to operate efficiency, performance, drivability and uptime – as a 6x4 at startup, during backup maneuvering or in attributes essential to customers. For line-haul other environments where traction is cruicial. As the applications, transmission and axles are typically truck nears a predetermined speed or condition, the selected such that the engine operates in its most inter-axle shaft disconnects from the power divider in efficient point at cruise conditions. Figure 8 shows the forward axle as well as the ring gear in the rear iso-contours of engine efficiency as a function of axle, allowing the axle to operate in a more efficient speed and power. The figure also illustrates where 6x2 mode. At the same time, it shifts the forward axle the engine operates while the truck is cruising at 65 to a faster ratio that enables engine downspeeding. miles per hour in the 11th and 12th gear, respectively. The Meritor detachable and liftable rear tandem axle We note that the gear selection has a significant concept [4] is another concept allowing impact on engine efficiency. interchangeable 6x2 / 6x4 operation. The overall powertrain efficiency is described by the following expression:

휂푃푡푟푎푖푛 = 휂퐸푛푔푖푛푒 ∙ 휂푇푟푎푛푠 ∙ 휂퐴푥푙푒 (1)

where 휂푃푇푟푎푖푛 is the overall powertrain efficiency, 휂퐸푛푔푖푛푒 represents the engine efficiency, 휂푇푟푎푛푠 is the

transmission efficiency and 휂퐴푥푙푒 is the axle efficiency. Figure 6 Examples of truck axle configurations Both transmission and axle efficiency generally Engine downspeeding is typically accompanied by improve with downspeeding, whereas there is an taller (lower numeric ratio) rear axles, which helps optimum for the engine as seen from Figure 8 and improve efficiency. However, it is important to also discussed in the prior section on engine technologies. consider the impact a taller ratio rear axle has on vehicle performance, NVH, driveline weight and torque management.

The recently introduced PACCAR 40,000 lbs. tandem drive axle, shown in Figure 7, offers gear ratios from 2.47 to 3.70 and supports both direct and overdrive drivetrains.

It features a patented through-shaft pinion design in order to eliminate gears in the forward drive axle and reduces oil churning losses with a unique passive lubrication system and the use of laser welded joints instead of bolts.

122 ICPC 2019 – 6.1

Figure 9 Suboptimal engine-transmission interaction

Figure 8 Engine efficiency vs. speed and region of In contrast, we consider an engine calibrated to operation at 65 mph cruise condition provide constant power across the gears, as illustrated in Figure 10. Similar to the case discussed The choice of engine rating is critical in achieving the above, the vehicle speed will decrease from the 65 overall desired powertrain performance and fuel mph set speed down to 61 mph as the 2% grade is economy. This is illustrated in the following by encountered. However, unlike the situation described contrasting the performance of two different engine in Figure 9, no downshift will occur since there is no ratings. Figure 9 shows the power available in the two more power available in 11th gear and the speed will top gears (11th and 12th) for a given application along remain at 61 mph with the transmission in 12th gear. with road loads corresponding to 0% (flat), 1%, 2%, The result is a slower ascent up the hill, but with less and 3% grade. The 12th gear represents an Over shifting and better fuel economy. Moreover, the Drive, enabling engine downspeeding. We imagine a reduced vehicle speed turns out to be advantageous scenario in which the truck is cruising at 65 mph in upon cresting the hill since it reduces the need for 12th gear as it approaches a 2% grade. Power will be engine braking and subsequent downhill speed increased until it reaches point A in the figure, at control. The net effect is typically a fairly minor impact which point no more power is available in the 12th on overall trip time but a noticeable positive impact on gear. Since insufficient power is available for the fuel economy. given grade, vehicle speed will start to decrease (transition from point A to B). When the vehicle speed hits the lower threshold of the cruise control speed window (assumed here to be 61 mph), a downshift to 11th gear is executed (B -> C). In 11th gear sufficient power is available to accelerate the vehicle back to the 65 mph set speed (C -> D), at which point the transmission performs an upshift in order to take advantage of the efficiency benefits of engine downspeeding (see Figure 8). The cycle is then repeated. The result is fluctuating vehicle speed and excessive shifting, which has a detrimental impact on fuel economy4.

Figure 10 Constant power engine rating

Predictive Functionality Vehicle connectivity has emerged as an important enabler to transport efficiency. Today’s trucks have

4 The fuel consumed by accelerating the engine speed during a downshift is never recovered.

123 ICPC 2019 – 6.1 access to GPS-based terrain information that makes it possible to control and optimize powertrain Automation and Connectivity performance. Examples of these so-called Advanced Predictive Features are: Our society is more connected than ever and through our smartphones and tablets there is a wealth of • Predictive Cruise Control information at our fingertips. As it relates to • Predictive Shifting commercial trucks, there is a tremendous amount of • Neutral Coast data being generated and transmitted between vehicles and infrastructure elements. This data can Figure 11 illustrates how these features operate in a include vehicle operating parameters, weather, traffic scenario where a truck is moving through a rolling hill information and logistical information communicated terrain. Predictive Cruise Control (PCC) maintains a by fleet operators and dealerships. The availability of vehicle speed set by the driver, but allows variation all this data offers a range of opportunities for new within a certain range to ensure optimal efficiency and products and services aimed at increasing uptime, drivability. As the truck approaches a hill, the productivity and customer satisfaction. The prior transmission executes a predictive downshift to discussion on predictive functionality is an example of ensure sufficient torque is available to ascend the how vehicle connectivity can be leveraged to improve grade. As the terrain plateaus, the transmission efficiency, but it is fair to say that the full potential of avoids a temporary upshift based on information that vehicle connectivity has yet to be realized. the vehicle is about to encounter another grade. As the truck climbs the hill, the PCC set speed is allowed Additional opportunities related to connectivity to droop knowing that speed will recover on the include smart routing that takes parameters such as downhill. Such a strategy results in less frequent weather and congestion into account in chartering the shifts and improves fuel economy without penalizing most efficient route for a truck. There is also overall trip time. Upon cresting the hill a feature significant progress made in the area of vehicle self- referred to as Neutral Coast is activated, which diagnostics, recognizing when failures are about to causes the transmission to shift into neutral with the occur and communicating to the nearest dealership engine idling. Future extensions of this strategy may what parts may be needed. Such predictive involve turning the engine off during the downhill maintenance will be key to maximizing truck uptime. coast event. This will obviously require uninterrupted power supply to critical accessories, such as power Connectivity is also a critical enabler to automation, steering. which is a topic receiving plenty of attention currently. As disruptive as many of the autonomous developments may seem, it is worth noting that automation really represents the continuation of a long-term trend. Many of the systems we today take for granted on vehicles, e.g. Anti-Lock Braking System (ABS), Electronic Stability Control (ESP), Automated Manual Transmissions (AMTs), manage tasks that originally fell on the driver to execute. A truck driving itself is obviously a significant extrapolation of that paradigm.

Advanced Driver Assistance Systems (ADAS) are aimed at improving safety, productivity and comfort of Figure 11 Advanced Predictive Features the driver. The SAE standard J3016 defines 5 ADAS levels, ranging from longitudinal or lateral control (e.g. FUTURE TECHNOLOGIES Adaptive Cruise Control, Lane Keeping Assist) to fully autonomous operation. The latter requires a large Key drivers to future technology development amount of additional sensors on the truck (cameras, continue to be Total Cost of Ownership (TCO), radars, LIDAR) and it is important to consider the performance and productivity, while meeting impact on TCO in deploying higher levels of increasingly stringent regulatory requirements. The automation. latter include the Greenhouse Gas Phase 2, the Low NOx regulation proposed by the California Air Truck platooning is an ADAS technology that has the Resources Board (CARB), emerging zero emission potential of improving fuel economy by allowing standards and urban area diesel bans. Technology trucks to draft. Platooning can be an ADAS level 1 or advancements will benefit from progress made in the 2 depending on whether the following truck(s) feature areas of automation, connectivity and electrification. either or both longitudinal and lateral control. Trucks

124 ICPC 2019 – 6.1 in a platoon typically communicate through a 5.9 GHz The key components of an all-electric powertrain are: Dedicated Short Range Communication (DSRC) battery pack, electric motor, inverter and potentially a protocol. Not having a driver in the loop puts high transmission depending on the application. The demands on the level of integration of throttle, brakes battery pack has to be sized to provide the desired and steering system. There are also serious range and have proper controls and thermal requirements regarding functional safety and system management to achieve adequate cycle life. redundancy. The voltage level is primarily dictated by the amount Electrification and Hybridization of power required, since the maximum current is effectively limited by the cost and weight of Several factors will drive increased adoption of hybrid connectors and wires as well as the ability to route the and electrification technologies in the commercial cables. Higher current levels are also associated with segment: winding losses.

• A growing number of cities and municipalities A key architecture decision relative to all-electric are in the process of implementing zero powertrains is where to place the electric motor. emission zones [5],[6] There are broadly speaking two possibilities: Central • Batteries have become more powerful, Drive and e-axle. The two are illustrated in Figure 12. reliable and less expensive There are pros and cons associated with each. An • The automotive segment has started to integrated e-axle benefits from a direct connection embrace electrification in a serious way between motor, transmission and wheel ends, not requiring a bevel gear to connect a driveline to the Each application - whether it be Pick-up & Delivery, rear axle. As such, e-Axle are a little bit more efficient refuse, regional-haul or long-haul - has a a different than the central drive topology. However, challenges set of requirements and an electrification concept that include unsprung masses, vibration, hose- and wire works for one application does not necessarily flexing and packaging (especially when a gearbox is translate to another. A key industry priority is to look necessary). E-axles are also not compatible with the for synergies, commonality and modularity in order to driveline and axle offerings today. accomplish efficient product development and attractive volumes of scale.

There are different degrees of electrification, ranging from all-electric powertrains to hybrids that rely on a separate powerplant5 for propulsion. Recognizing the breadth of the topic, the current section will be limited to a discussion on all-electric applications and a mild hybrid featuring accessory electrification.

Absent regulation requiring zero emission operation, there is no strong business case for an all-electric Class 8 long-haul truck. Given today’s battery technology, the electric equivalent of a line-haul truck carrying two 80-gallon tanks of diesel would have to Figure 12 Electric Powertrain Topologies carry approximately 25,000 kg of batteries6, far offsetting the weight benefit of removing the engine, Another key design decision is whether or not to aftertreatment and transmission. At roughly connect a transmission to the electric motor. While $200/kWh for typical state-of-the-art Li-ion less speed sensitive than diesel engines, the technology, this would result in a cost upwards of efficiency of electric motors does vary with speed and $500,000 just in batteries. For Medium-Duty there are limits to their operation. The e-motor ratio applications with their more limited range and power defines the ratio between maximum and nominal requirements, there is a more attractive and nearer motor speed. Depending on this ratio and the term business case. Incidentally, these are also the application requirements, a 1-4 speed transmission type of applications that are likely to be most exposed may be necessary. The most common motor types to zero emission regulation introduced in urban areas. are Permanent Magnet Synchronous Machines

5 In most cases this power plant is an Internal Combustion Engine, 6 Assumes 8 MPG for a diesel powered truck, 2 kWh/mile and a although recently fuel cells are receiving renewed attention. battery weight of 10 kg/kWh.

125 ICPC 2019 – 6.1

(PMSM), Induction Magnet (IM), Synchro Reluctance An onboard 48V electrical system also opens up (SynRel), and External Excited Synchro Machine other interesting possibilities. The Low NOx regulation (EESM). Of these the PMSM and the EESM typically proposed by CARB for 2023-24 will likely require offer the highest efficiency and power density, wheras some kind of thermal management of the exhaust the other types are more attractive in terms of cost. gas. This is something that could be accomplished by incorporating a 48V electrical heater. A 48V mild hybrid offers an interesting intermediate step between the conventional powertrains of today CONCLUSION and tomorrow’s all-electric ones. A key attribute of the 48V mild hybrid is the ability to power accessories, Commercial trucking is at the heart of the such as power steering, AC compressor and water transportation sector and a cruicial contributor to the pump independent from the engine. Such “smart” growth of the global economy. Decades of continuous operation of accessories can yield notable efficiency improvement in engine, transmission and overall benefits. It is also possible to replace the 12V starter powertrain technology have made trucks operating on with a more robust 48V system. Similar to the all- the roads today cleaner and more efficient than ever electric topology, a key design decision for the mild before. Powertrain integration has been a key enabler hybrid is the position of the electric motor. A Belt to this by focusing on overall system optimization as integrated Starter Generator (BiSG) is a common and opposed to attempting to maximize the performance cost effective mild hybrid topology owing to its of individual subsystems. relatively minor impact on the existing vehicle architecture. However, this architecture does not Connectivity has emerged as an enabler to improving support coasting. Figure 13 shows an alternative truck efficiency and uptime through predictive approach where a 48V electric motor is connected to functionality and diagnostics. Continued automation, a transmission mounted Power Take-Off (PTO) unit. up to and including autonomous operation, is already The PTO unit is also connected to the AC having a significant impact on the trucking industry compressor, thus eliminating the one on the belt. This and this trend is likely to continue. architecture supports removal of the engine Front End Accessory Drive (FEAD), which frees up space Future regulatory and competitive pressures will under the hood and enables optimization of the require solutions for zero emission operation in urban aerodynamic performance. and non-attainment areas, which is spurring development efforts in the realm of electrification. A key challenge for the OEMs is to identify a modular and scalable architecture that can cover all relevant applications. Battery technology development is progressing rapidly, but there is still not a compelling business case for Heavy-Duty all-electric long-haul trucks in the absence of incentives or zero emission mandates. Medium Duty battery electric applications appear more feasible in the near term and a 48V mild hybrid offers an attractive stepping stone between today’s conventional powertrain and future all-electric configurations.

Figure 13 48V mild hybrid with accessory REFERENCES electrification “American Trucking Trends 2018”, American During normal cruise operation, the engine is Trucking Association, 2018 connected to the transmission, which provides DOE Vehicle Technologies Office Annual mechanical energy to the drivetrain and drives the Progress Report 2018 accessories through the PTO unit. At the same time Dana e-Newsletter, Issue 3, 2015 the batteries are being charged. Meritor product information, 2018 When the truck is coasting down a hill, the engine can “The Zero Emission Vehicle (ZEV) Regulation”, be turned off while power is still provided to all CARB Fact Sheet, August 2018 accessories. At the same time, the batteries can be “City bans are spreading in Europe”, Briefing by charged with excess energy available through Transport & Environment, October 2018 regenerative braking.

126 ICPC 2019 – 6.1

DEFINITIONS, ACRONYMS, ABBREVIATIONS

ABS Anti-Lock Braking ADAS Advanced Driver Assistance Systems AMT Automated Manual Transmission ASC Ammonia Slip Catalyst BiSG Belt integrated Starter Generator BTE Brake Thermal Efficiency CARB California Air Resources Board DEF Diesel Exhaust Fluid DD Direct Drive DOC Diesel Oxidation Catalyst DPF Diesel Particulate Filter DSRC Dedicated Short Range Communication EGR Exhaust Gas Recirculation ESC Electronic Stability Control NOx Nitrogen Oxides NVH Noise, Vibration and Harshness OBD On-Board Diagnostics OD Over Drive ORC Organic Rankine Cycle OTR Over The Road PM Particulate Matter PTO Power Take-Off SCR Selective Catalytic Reduction TCO Total Cost of Ownership VGT Variable Geometry Turbocharger WHR Waste Heat Recovery

127

ICPC 2019 – 6.2 Value Creation in the Commercial Powertrain Industry

Dr. Albert Neumann, Jana Mühlig, Xaver Müller Strategy Engineers

Copyright © 2019 AVL List GmbH, Strategy Engineers and SAE International

ABSTRACT future in the commercial vehicle (CV) industry will develop. The commercial vehicle industry is about to change significantly in the upcoming years. While the INTRODUCTION traditional diesel dominated world will still generate main revenues and profits, future value creation In the past years, the global sales for commercial needs to be based on new technologies and vehicles (comprising of light-duty, medium-duty, businesses. Electrification, Connectivity and heavy-duty trucks and city busses in the course of this Autonomous Driving can all be part of this value paper) have been growing steadily. But this growth creation but not without risks as major investments will slow down when looking at published forecast are required to create the foundation for a profitable data (compare Exhibit 1). implementation. Back in the year 2000, the truck market was growing In this paper, we discuss how the traditional world will at a compound annual growth rate (CAGR) of almost develop and how operational improvements are 5%. This growth was interrupted in the late 2000’s by mandatory to maintain current profits. The main focus the financial crisis. Since recovery, the market has is on reviewing challenges and opportunities for the grown at a CAGR of over 2% until today. Looking new technologies. For the three trends Alternative forward, we expect future transportation demand to Powertrains, Connectivity and System Solutions as be still growing but at low rate of approximately less well as Autonomous Driving, we present our view of than 2% annually. major risks and opportunities and how we believe the Future Today +1.8% Before financial crisis +2.1% +4.5%

2000 2005 2010 2015 2020 2025 2030 CAGR City bus Heavy-duty truck (> 11.8 t) Medium-duty truck (< 11.8 t) Light-duty truck (<3.5 t)

Exhibit 1: commercial vehicles sales volume in units, worldwide, based on [1]

128 ICPC 2019 – 6.2

Looking towards 2030, the profit structure of The city of London serves as a practical example in commercial vehicle companies can be divided into setting strict targets to reduce the local transport two key areas: traditional business and new emissions in the city. In the capital of the United opportunities (compare Exhibit 2). Kingdom, the government is gradually installing and aggravating Ultra Low Emission Zone (ULEZ) and In the first area of traditional business, we see a price Low Emission Zones (LEZ). In those zones, only pressure across the industry due to lower growth vehicles complying with strictly defined emission rates and through new competitors which will lower standards are allowed, otherwise heavy fines are due margins and hence generate less profits. At the same - especially for commercial vehicles. Regarding the time, commercial vehicle companies are and will city bus fleet, the city’s target is to have a fully zero continue to improve their operational excellence, emission fleet by 2037. An investment of £300 million such as optimization of product costs and internal is planned by the mayor of London to transform the processes. bus fleet including retrofitting thousands of buses and phasing out pure diesel double-deck buses from 2018 New opportunities onwards. [3] (focus of this article) London only serves as one practical example. In fact, almost every larger city in the world is aiming at Traditional business Driving improving their air quality. Especially city busses and

Autonomous Autonomous light-duty trucks that operate in or close to the city

centre are expected to be influenced by these

Connectivity and and Connectivity Alternative Alternative

System Solutions System developments.

Powertrains

Excellence

Market and and Market Operational Operational Besides restricted emission zones, overall CO2 Macroeconomics efficiency of diesel engines is becoming increasingly Today 2030 challenging and other alternative powertrains (hydrogen/fuel cell, hybrid, synthetic fuels, biofuels) Exhibit 2: Profit structure of commercial vehicle are likely to gain importance in achieving emission companies (illustrative), in EUR, based on [2] goals. Exhibit 3 shows our forecast on the upcoming distribution of alternative powertrains by vehicle type. All in all, we believe that commercial vehicle companies are capable of outperforming the external City bus LD truck factors and increase their profits based on the (<3.5 t1) operational excellence improvement. 168k 7.4M 23% 14% Besides the traditional business, we see – and this is 1% 13% 7% 28% what this paper is all about – new opportunities which 3% have the potential to significantly improve overall profits. The areas in which these profits can be obtained are:

I - Alternative Powertrains 2018 2030 2018 2030 MD truck HD truck II - Connectivity and System Solutions (< 11.8 t1) (>11.8 t1)

III - Autonomous Driving 1.0M 11k 2% 23% 0.3% 0.3% 0.1% In the following chapters, these three trends are covered in more detail. Technology trends, practical examples arising opportunities but also challenges are discussed. I - ALTERNATIVE POWERTRAINS 2018 2030 2018 2030 As seen in the passenger car industry, the main driver Diesel / Petrol Hybrid Electric & Fuel Cell for alternative powertrains in the commercial vehicle Exhibit 3: Powertrain forecast for commercial industry are legislative frameworks and zero emission vehicles, share of new sales (volume) worldwide [1] zones.

129 ICPC 2019 – 6.2

The application of alternative powertrains depends on creation. Along with this, it is essential to define a new use case and timing. We see battery electric vehicles set of modular architectures that can be applied to be the superior technology in distribution and city- across different applications. In the field of centric use cases. Fuel cell propulsion will be a strong production, companies have to verify their make-or- contender as it is best suited for heavy-duty transport buy strategy as some components may be segments that require long haul range and high outsourced and others will be produced inhouse for utilization. Main factors are higher energy density, competitive advantage. This also requires setting up lower weight and faster re-fuelling. The success of a new supplier base. On the operational level, CNG/LNG as well as synthetic fuels and biofuels will standardized hardware and software interfaces as heavily depend on market price. This is subject to the well as an easy access to charging infrastructure availability of renewable energies and resources. need to be established. Furthermore, aftersales support to customers is key as these new Now, the main question is, how this can lead to more technologies may require additional accessories such profit for CV companies. The answer is to be found in as intelligent switchboards or battery monitoring. the different use cases and legislation restrictions, making these technologies more or less attractive for customers. In any case, the overall TCO and/or usability of the product will make the difference. Then, Gain new E/E competencies it is important that despite increasing material and Define new electrical product platforms

product cost, the offer to the client is attractive. Development

We strongly believe that there is an opportunity to Verify their make-or-buy decisions charge the customer for these new technologies. Set up a new supplier base Because of this, CV companies are still able to Restructure their manufacturing generate add-on profit from alternative powertrains capabilities despite higher material costs. Production

As shown in Exhibit 4, we expect the additional value Standardizehardware and software creation to be approximately 5.7% until 2030. interfaces

Broaden their infrastructure Operation Battery electric drive +5.7% +9.4% 12% components 4% Exhibit 5: Key challenges in the area of Alternative Powertrains

Combustion engine 72% -0.5% 64% II - CONNECTIVITY AND SYSTEM components SOLUTIONS

The next big area with great potential to increase the Gearbox and profit potential in upcoming years is what we 24% +0.5% 24% differential generally name as connectivity. This can be services

2018 2030 very close to the traditional business and products such as applications for fleet operators to manage, CAGR maintain and optimize their fleet, but can also reach Exhibit 4: Expected value creation in propulsion to areas beyond the core business into service technologies of commercial vehicle companies, in portfolio extensions with data-based business EUR share [4] solutions. The competition in these areas is quite divers as new players are moving in, trying to capture As already indicated, a number of challenges have to part of the overall future value chain of transport (see be met in order to achieve this additional profit. The Exhibit 6). key challenges can be clustered along three different areas and cover the entire value chain from A great example for a connectivity platform is RIO, a development to production and operations (see digital brand established by the TRATON Group in Exhibit 5). 2017. Companies like Scania, MAN, Continental, Meiller Kipper or Schmitz Cargobull are already using Due to the fundamentally different technologies, CV the open connectivity platform to manage their companies must build up new E/E competencies vehicle fleets. In form of a box, RIO can be retrofitted required for electric-driven vehicles. Development to trucks from every manufacturer and can be used to competencies in e.g. batteries, e-motors, inverters or control single trucks of the fleet, to determine the power electronics is key to successful product vehicles’ positioning, for intelligent routing and route

130 ICPC 2019 – 6.2 analysis, to plan and optimize maintenance of the The first stage are vehicle-integrated services, where whole fleet and to analyse performance of the the value add relates to the vehicle itself. This vehicles. includes basic features such as navigation, infotainment or fuel management functions. The next stage is to offer a commercial vehicle related portfolio of services that benefits the customer retention. This includes e.g. fleet management, maintenance, driver new analysis or digital payment functions. All the functions CV companies of the first two stages are mainly standard and part of the portfolio of all companies - there is very little differentiation potential. The objective for any CV company should be to develop and market services that go beyond the vehicle and cover additional elements of the transport value chain. This can be traditional CV companies applications for load capacity sharing, truck sharing, aftersales or other digital services offered to transport operators.

To develop, implement and manage digital services, fleet CV companies must manage a vastly different tier I management suppliers business compared to today’s core business. It providers requires different people with different skills, methods and processes (see Exhibit 8). telematics logistics tech service providers giants providers Shorten their product development cycles Set-up agile workingmethods Gain SW development knowledge Establish partnering models

Development Think customer-centric Exhibit 6: Players in the connected commercial vehicle market, non-exhaustive exemplary selection Develop new supplier management This is only one example and we expect 80-90% of Enable on-/offboard computing

the global fleet to be “connected” in 2030, at least in Offer flexible/efficient updatability Production Europe and NAFTA. The challenge is how to use connectivity for new or extended business models and hence increase the overall profit for CV Re-positionwithin the value chain companies. Build-up reliable data protection

Extend customer contact Operation There are three different areas of services that we see on the market, based on their proximity to the commercial vehicle (see Exhibit 7). Exhibit 8: Key challenges in the area of Connectivity and System Solutions

The development of any digital service requires in- depth customer understanding and fast realization. Application and software development capabilities are key to be successful. Developing software calls for a new and independent software development process with shorter product development cycles and continuous product updates. Connected with that, agile working methods have to be set up and clear synchronisation points between the traditional hardware based and software development have to be defined.

Figure 7: Value adding services for connected Connected vehicles will produce a significant amount commercial vehicles of data that a future E/E architecture needs to cope with. Data needs to be handled onboard within the car

131 ICPC 2019 – 6.2 but also offboard, as cloud-based applications or on unmanned trucks by 2022-2023, but also intends to mobile computing devices. use fuel cell powertrains in their vehicles.

Last but not least, a whole new concept for data As shown in Exhibit 9, we consider that autonomous security is necessary to manage digital services driving will develop in different phases and that fully connected to vehicles or outside the vehicle unmanned operation in all use cases will only be perimeter. available far after 2030.

III - AUTONOMOUS DRIVING Autonomous trucks promise profit potential for OEMs as well as TCO optimization potential for their Autonomous driving aims to enable unmanned customers (see Exhibit 10). Main savings are driving of trucks and will heavily influence the truck expected from unmanned driving, but also through business in the future. more fuel efficiency. Depending on the use case, this can sum up to 30% of TCO savings for the vehicle TuSimple, a California-based start-up, already owners. Additionally, the annual mileage can be more operates 12 autonomous trucks on Arizonan streets than doubled due to higher utilization potential of and aims to increase this number to 50 until June driverless trucks. 2019. The company which was founded in 2015 and operates from offices in San Diego and Peking raised TCO composition Annual mileage in thousand km per year per HD truck over USD 178.1 million in four funding rounds and in thousand EUR per HD truck therefor gained unicorn status [5]. The Chief Product 800 291 Driver Officer Chuck Price mentioned that they “are -30% 560 confident that [they] will have [their] first commercial Fuel x 2.6 113 driverless operation in late 2020 to 2021” [6]. For Acquisition safety reasons, their trucks are still manned with a system engineer and a driver but basically drive Other autonomously. The company just managed to get a Non- Autonomous Non- Autonomous public road test licence for Shanghai and by autonomous HD truck autonomous HD truck HD truck HD truck developing a new night camera, they accomplished to raise the daily utilization from 12 hours (50%) to 19 hours (80%) [7]. Exhibit 10: Optimization potential through autonomous trucks, own calculation based on [9] and Another example from the conventional commercial [10] vehicle industry is Volvo trucks which operates six autonomous mining trucks in a Norwegian limestone To succeed in autonomous driving technologies, extraction mine. The trucks from the type Volvo FH commercial vehicle companies have to face different 16 are controlled by a wheel loader operator using a challenges, as shown in Exhibit 11. site management system and are currently in test mode. Volvo trucks plans to operate the autonomous The key challenge is the development, validation and trucks 24/7 autonomous from autumn 2019 on [8]. homologation of assisted and autonomous driving Another innovator is Nikola, a US based start-up functionalities. This not only requires skilled software company which not only announced fully autonomous developers and capable partners for both development and validation, but also a new E/E Fully autonomous driving Autonomous trucks will first be without geo restrictions established in geo fenced, confined areas far after 2030

Partial High automation in High automation on Full automation without automation confined areas selected highways geo restrictions

2015 2020 2025 2030 unmanned operation in all use cases Exhibit 9: Roadmap of autonomous commercial vehicles

132 ICPC 2019 – 6.2 architecture with high performance processors. In New competencies are required throughout all three contrast to mechanical architectures, E/E growth fields. Synergies can be drawn from building- architectures can be cost-efficiently used across up E/E know-how, promoting software development various derivatives. skills and respective methods and enhance manufacturing capabilities.

Set-up an agile SW development approach Develop a flexible, modular E/E

architecture New value chains Development

Harmonize legal frameworks New business models Establish partnershipswith tech companies

Production New competencies

Clarify TCO benefits New vehicle platforms Build effective cyber security

Monetize their autonomous features Operation Exhibit 12: Main tasks for commercial vehicle companies Exhibit 11: Key challenges in the area of Autonomous Driving Changing vehicle platforms move away from mechanical-based platforms to electrical-driven CONCLUSION platforms and require modular E/E architectures as well as standardized interfaces of hard- and software. Commercial vehicle companies are facing a challenging time with traditional markets still growing REFERENCES but at lower pace. Maintaining current profits is key and can only be managed with continuous [1] IHS Markit Vehicle Industry Forecast, 2018 improvement of operational excellence. [2] McKinsey&Company: “Route 2030 – the fast track to the future of the commercial vehicle New technologies and business development offer a industry”, great chance to add additional profits to the existing September 2018 core business. However, this requires a significant [3] Mayor of London / London Assembly – Transport: change in all areas of a company. “Green Transport”, 2019 We see four main tasks for CV companies: New value [4] VDMA, Forum Elektromobilität: “Antrieb im chains, new business models, new competencies and Wandel”, March 2018 new vehicle platforms. All four have to be tackled [5] TuSimple Media page via Forbes: “First Robo- simultaneously in order to succeed in Alternative Trucking Unicorn? TuSimple Delivers $95 Million Powertrains, Connectivity and System Solutions and Funding Round” by Alan Ohnsmann, February Autonomous Driving (see Exhibit 12). 13, 2019 [6] The Wall Street Journal, Logistics Report: “Self- Transforming value chains require the CV companies Driving Truck Tech Startup TuSimple Raises $95 to re-think their value chain positioning in forms of e.g. Million in New Funding” by Jennifer Smith, vertical integration. They also need to reconsider their February 13, 2019 make-or-buy strategy when it comes to [7] TuSimple Media page via VentureBeat: manufacturing of hardware components. A changing “Driverless trucks ride at night with TuSimple’s supplier base will also require new partnering models improved camera system” by Seth Colaner, through the three fields. March 19, 2019 Emerging business models call for a change of [8] Volvo Trucks Magazine, Business Story: mindset inside commercial vehicle companies. They “Autonomous trucks in real operation” by Alastair have to implement an expanded service offering Macduff, February 20, 2019 portfolio and tackle product digitalisation under close [9] Statista.de consideration of their customer (customer centrism).

133 ICPC 2019 – 6.2

[10] “Analysis of long haul battery electric trucks in EU” by Earl, Mathieu, Cornelis et al., originally presented in “8th Commercial Vehicle Workshop, Graz”, May 17-18, 2018

DEFINITIONS, ACRONYMS, ABBREVIATIONS

CAGR Compound Annual Growth Rate CV Commercial vehicle E/E Electrics/Electronics OEM Original Equipment Manufacturer TCO Total Cost of Ownership

134