Advances in Intelligent Systems and Computing 442

Mohammad S. Obaidat Janusz Kacprzyk Tuncer Ören Joaquim Filipe Editors Simulation and Modeling Methodologies, Technologies and Applications International Conference, SIMULTECH 2015 Colmar, France, July 21–23, 2015 Revised Selected Papers Advances in Intelligent Systems and Computing

Volume 442

Series editor Janusz Kacprzyk, Polish Academy of Sciences, Warsaw, Poland e-mail: [email protected] About this Series

The series “Advances in Intelligent Systems and Computing” contains publications on theory, applications, and design methods of Intelligent Systems and Intelligent Computing. Virtually all disciplines such as engineering, natural sciences, computer and information science, ICT, economics, business, e-commerce, environment, healthcare, life science are covered. The list of topics spans all the areas of modern intelligent systems and computing. The publications within “Advances in Intelligent Systems and Computing” are primarily textbooks and proceedings of important conferences, symposia and congresses. They cover significant recent developments in the field, both of a foundational and applicable character. An important characteristic feature of the series is the short publication time and world-wide distribution. This permits a rapid and broad dissemination of research results. Advisory Board

Chairman Nikhil R. Pal, Indian Statistical Institute, Kolkata, India e-mail: [email protected] Members Rafael Bello, Universidad Central “Marta Abreu” de Las Villas, Santa Clara, Cuba e-mail: [email protected] Emilio S. Corchado, of Salamanca, Salamanca, e-mail: [email protected] Hani Hagras, University of Essex, Colchester, UK e-mail: [email protected] László T. Kóczy, Széchenyi István University, Győr, Hungary e-mail: [email protected] Vladik Kreinovich, University of Texas at El Paso, El Paso, USA e-mail: [email protected] Chin-Teng Lin, National Chiao Tung University, Hsinchu, Taiwan e-mail: [email protected] Jie Lu, University of Technology, Sydney, Australia e-mail: [email protected] Patricia Melin, Tijuana Institute of Technology, Tijuana, Mexico e-mail: [email protected] Nadia Nedjah, State University of Rio de Janeiro, Rio de Janeiro, Brazil e-mail: [email protected] Ngoc Thanh Nguyen, Wroclaw University of Technology, Wroclaw, Poland e-mail: [email protected] Jun Wang, The Chinese University of Hong Kong, Shatin, Hong Kong e-mail: [email protected]

More information about this series at http://www.springer.com/series/11156 Mohammad S. Obaidat • Janusz Kacprzyk Tuncer Ören • Joaquim Filipe Editors

Simulation and Modeling Methodologies, Technologies and Applications International Conference, SIMULTECH 2015 Colmar, France, July 21–23, 2015 Revised Selected Papers

123 Editors Mohammad S. Obaidat Tuncer Ören Department of Computer and Information Faculty of Engineering, School of Electrical Science Engineering and Computer Science Fordham University University of Ottawa Bronx, NY Ottawa, ON USA Canada

Janusz Kacprzyk Joaquim Filipe Systems Research Institute Department of Systems and Informatics Polish Academy of Sciences Polytechnic Institute of Setúbal Warsaw Setúbal Poland Portugal

ISSN 2194-5357 ISSN 2194-5365 (electronic) Advances in Intelligent Systems and Computing ISBN 978-3-319-31294-1 ISBN 978-3-319-31295-8 (eBook) DOI 10.1007/978-3-319-31295-8

Library of Congress Control Number: 2015955364

© Springer International Publishing Switzerland 2016 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made.

Printed on acid-free paper

This Springer imprint is published by Springer Nature The registered company is Springer International Publishing AG Switzerland Preface

This book includes extended and revised versions of a set of selected papers from the 5th International Conference on Simulation and Modeling Methodologies, Technologies and Applications (SIMULTECH 2015), which is sponsored by the Institute for Systems and Technologies of Information, Control and Communication (INSTICC) and co-organized by the University of Haute Alsace. SIMULTECH is held in cooperation with the ACM SIGSIM—Special Interest Group (SIG) on SImulation and Modeling (SIM), Japan Society for Simulation Technology (JSST), Federation of Asia Simulation Societies (ASIASIM), AIS Special Interest Group on Modeling and Simulation (AIS SIGMAS), Federation of European Simulation Societies (EUROSIM), Movimento Italiano Modellazione e Simulazione (MIMOS), and European Council for Modelling and Simulation (ECMS). It is also technically co-sponsored by Liophant Simulation and Simulation Team. This conference brings together researchers, engineers, and practitioners inter- ested in methodologies and applications of modeling and simulation. The main topics covered in the papers accepted in the conference are: Methodologies, tech- nologies, tools, and applications of modeling and simulation. The carefully selected and accepted papers demonstrate new and innovative solutions. They also highlight technical issues and challenges in this field. SIMULTECH 2015 received 102 paper submissions, including special sessions, from 36 countries in all continents, of which 53 % were orally presented (19 % as full papers). A double-blind peer paper review was performed by the international Program Committee members, all of them recognized in at least one of the main conference topic areas. The papers included in this book were selected from those with the best reviews taking also into account the quality of their presentation at the conference, assessed by the session chairs. Therefore, we hope that you find the papers included in this book interesting, and we trust they will represent as helpful references. We wish to thank all those who supported and helped to organize the conference. On behalf of the conference Organizing Committee, we would like to thank the authors whose work mostly contributed to a very successful conference; the

v vi Preface members of the Program Committee whose expertise and diligence were instru- mental to ensure the quality of final contributions; and the invited distinguished keynote speakers for their invaluable contribution. We also like to thank the dis- tinguished Panelists who set up an energizing tone and made excellent remarks that were instrumental in making the conference more interesting and valuable to attendees. We also wish to thank all the members of the Organizing Committee whose work and commitment were invaluable. Thanks to the administrators, faculty, staff and students of the University of Haute-Alsace, France especially Prof. Pascal Lorenz for hosting the conference and for their great help in local arrangements. Thanks also are due to the organizations that technically co-sponsored the confer- ence. Special thanks go to the Staff members of SIMULTECH 2015 for their outstanding support and dedicated work. Last but not least, we would like to thank INSTICC for sponsoring and organizing the conference.

July 2015 Mohammad S. Obaidat Janusz Kacprzyk Tuncer Ören Joaquim Filipe Organization

Conference Chair

Mohammad S. Obaidat, Fordham University, United States

Program Co-Chairs

Tuncer Ören, University of Ottawa, Canada Janusz Kacprzyk, Polish Academy of Sciences, Poland

Program Committee

Magdiel Ablan, Universidad de Los Andes, Venezuela Nael Abu-Ghazaleh, University of California, Riverside, United States Carole Adam, LIG-UJF, France Marco Aldinucci, University of Torino, Italy Mikulas Alexik, University of Zilina, Slovak Republic Manuel Alfonseca, Universidad Autonoma de Madrid, Spain Gianfranco Balbo, University of Torino, Italy Simonetta Balsamo, University of Venezia Ca’ Foscari, Italy M. Gisela Bardossy, University of Baltimore, United States Isaac Barjis, City University of New York, United States Jordi Mongay Batalla, National Institute of Telecommunications, Poland Nicolas Belloir, LIUPPA, France Lucian Bentea, University of Oslo, Norway Wolfgang Borutzky, Bonn-Rhein-Sieg University of Applied Sciences, Germany Christos Bouras, University of Patras and CTI&P Diophantus, Greece

vii viii Organization

Mario Bravetti, University if Bologna, Italy Hajo Broersma, University of Twente, The Netherlands Robin T. Bye, Aalesund University College, Norway Christian Callegari, University of Pisa, Italy Jesus Carretero, Computer Architecture Group, University Carlos III of Madrid, Spain Rodrigo Castro, University of Buenos Aires, Argentina Krzysztof Cetnarowicz, AGH—University of Science and Technology, Poland Srinivas Chakravarthy, Kettering University, United States Jiangzhuo Chen, Virginia Polytechnic Institute and State University, United States Lawrence Chung, The University of Texas at Dallas, United States Franco Cicirelli, Università della Calabria, Italy Tanja Clees, Fraunhofer Institute for Algorithms and Scientific Computing (SCAI), Germany Douglas Creighton, Deakin University, Australia Andrea D’Ambrogio, Università di Roma “Tor Vergata”, Italy Gabriella Dellino, National Research Council of Italy, Italy Atakan Dogan, Anadolu University, Turkey Zhihui Du, Tsinghua University, China Julie Dugdale, Laboratoire d’Informatique de , France Stephan Eidenbenz, Los Alamos National Laboratory, United States Sabeur Elkosantini, King Saud University, Saudi Arabia Zuhal Erden, ATILIM University, Turkey Denis Filatov, Centre for Computing Research (CIC), National Polytechnic Institute (IPN), Mexico Paul Fishwick, University of Texas at Dallas, United States Jason Friedman, Tel-Aviv University, Israel Richard Fujimoto, Georgia Institute of Technology, United States Marco Furini, Università di Modena e Reggio Emilia, Italy José Manuel Galán, Universidad de Burgos, Spain Petia Georgieva, University of Aveiro, Portugal Charlotte Gerritsen, Vrije Universiteit Amsterdam, The Netherlands Brian Goldiez, University of Central Florida, United States John (Yannis) Goulermas, The University of Liverpool, UK Alexandra Grancharova, University of Chemical Technology and Metallurgy, Bulgaria Francisco Grimaldo, Universitat de València, Spain Feng Gu, The City University of New York, United States Mykola Gusti, International Institute for Applied Systems Analysis, Austria Mirsad Hadzikadic, University of North Carolina Charlotte, United States Sigurdur Hafstein, Reykajvik University, Iceland Zhi Han, The MathWorks, United States Samer Hassan, Universidad Complutense de Madrid, Spain Cathal Heavey, University of Limerick, Ireland Monika Heiner, Brandenburg University of Technology Cottbus, Germany Organization ix

Tsan-sheng Hsu, Institute of Information Science, Academia Sinica, Taiwan Xiaolin Hu, Georgia State University, United States Ahmed Ibrahim, InterDigital Communications Inc, United States Eric Innocenti, IUT DE CORSE—University of Corsica, France Nobuaki Ishii, Bunkyo University, Japan Mhamed Itmi, INSA, Rouen, France Mura Ivan, University EAN, Colombia Yumi Iwashita, Kyushu University, Japan Luis Izquierdo, University of Burgos, Spain Syed Waqar ul Qounain Jaffry, University of the Punjab, Pakistan Mats Jägstam, University of Jönköping, Sweden Janusz Kacprzyk, Polish Academy of Sciences, Poland Cara Kahl, Hamburg University of Technology, Germany Anniken Karlsen, Aalesund University College, Norway Peter Kemper, College of William and Mary, United States William Knottenbelt, Imperial College , UK Juš Kocijan, Jozef Stefan Institute, Slovenia Petia Koprinkova-Hristova, Institute of Information and Communication Technologies, Bulgaria Sunil Kothari, HP Labs, United States Claudia Krull, Otto-von-Guericke University, Germany Pierre L’Ecuyer, Universite de Montreal, Canada Leifur Leifsson, Reykjavik University, Iceland Alberto Leva, Politecnico di Milano, Italy Fengyuan Li, Freescale Semiconductors, United States Jason Liu, Florida International University, United States Rong Liu, University of California, Los Angeles, United States Margaret Loper, Georgia Tech Research Institute, United States Antonio Mendes Lopes, University of Porto, Portugal Ulf Lotzmann, University of Koblenz, Germany Johannes Lüthi, FH Kufstein Tirol, Austria José Tenreiro Machado, Institute of Engineering, Polytechnic of Porto, Portugal Carla Martin-Villalba, UNED, Spain Moreno Marzolla, University of Bologna, Italy Radek Matušu, Tomas Bata University in Zlin, Czech Republic Roger McHaney, Kansas State University, United States Nuno Melão, Instituto Politécnico de Viseu, Escola Superior de Tecnologia e Gestão de Viseu, Portugal Adel Mhamdi, RWTH Aachen University, Germany Bozena Mielczarek, Wroclaw University of Technology, Poland Federico Milani, CHEM.CO Consultant, Italy Gabriele Milani, Politecnico di Milano, Italy Michael Möhring, University of Koblenz, Germany Roberto Montemanni, IDSIA—Dalle Molle Institute for Artificial Intelligence (USI-SUPSI), Switzerland x Organization

Jairo R. Montoya-Torres, Universidad de La Sabana, Colombia Tingting Mu, University of Liverpool, UK Navonil Mustafee, University of Exeter, UK Àngela Nebot, Universitat Politècnica de Catalunya, Spain Guyh Dituba Ngoma, Université du Québec en Abitibi-Témiscamingue, Canada Libero Nigro, Università della Calabria, Italy Lialia Nikitina, Fraunhofer Institute for Algorithms and Scientific Computing (SCAI), Germany Manuel Noguera, Universidad de Granada, Spain Michael J. North, Argonne National Laboratory, United States Paulo Novais, Universidade do Minho, Portugal James J. Nutaro, Oak Ridge National Laboratory, United States Sorin Olaru, SUPELEC, INRIA, France Paulo Moura Oliveira, Universidade de Tras os Montes e Alto Douro, Portugal Stephan Onggo, Lancaster University, UK Tuncer Ören, University of Ottawa, Canada Feng Pan, Liaoning Normal University, China Ioannis Paraskevopoulos, Anglia Ruskin University, UK George Pavlidis, “Athena” Research Centre, Greece Krzysztof Pawlikowski, University of Canterbury, New Zealand Ana Peleteiro, Universidade de Vigo, Spain L. Felipe Perrone, Bucknell University, United States Alexandre Petrenko, Centre de Recherche Informatique de Montreal, Canada Régis Plateaux, SUPMECA, France Veljko Potkonjak, University of , Serbia Tomas Potuzak, University of West Bohemia, Czech Republic Bhanu Prasad, Florida A&M University, United States Francesco Quaglia, Sapienza Università di Roma, Italy Francesc Josep Miguel Quesada, Universitat Autonoma de , Spain Martin Quinson, Université de Lorraine, France Jacinto A. Dávila Quintero, Universidad de Los Andes, Venezuela Manuel Resinas, Universidad de Sevilla, Spain M.R. Riazi, Kuwait University, Kuwait José Risco-Martín, Universidad Complutense de Madrid, Spain Ella E. Roubtsova, Open University of the Netherlands, The Netherlands Willem Hermanus le Roux, CSIR, South Africa Jordi Sabater-Mir, IIIA—CSIC, Spain Janos Sallai, Vanderbilt University, United States Paulo Salvador, Instituto de Telecomunicações/University of Aveiro, Portugal Antonio Sánchez-Ruiz, Universidad Complutense de Madrid, Spain María Teresa Parra Santos, University of , Spain Jean-François Santucci, SPE UMR CNRS 6134—University of Corsica, France Florence Sèdes, IRIT, France João Sequeira, Instituto Superior Técnico/Institute for Systems and Robotics, Portugal Organization xi

Philippe Serré, Supméca, France Clifford A. Shaffer, Virginia Tech, United States Flavio S. Correa Da Silva, University of Sao Paulo, Brazil Jaroslav Sklenar, University of Malta, Malta Andrzej Sluzek, Khalifa University, United Arab Emirates Jefrey Smith, Auburn University, United States James C. Spall, The Johns Hopkins University, United States Giovanni Stea, University of Pisa, Italy Mu-Chun Su, National Central University, Taiwan Nary Subramanian, University of Texas at Dallas, United States Peter Summons, University of Newcastle, Australia Antuela A. Tako, Loughborough University, UK Halina Tarasiuk, Warsaw University of Technology, Poland Pietro Terna, Università di Torino, Italy Mamadou K. Traoré, , France Klaus G. Troitzsch, University of Koblenz-Landau, Koblenz Campus, Germany Giuseppe A. Trunfio, University of Sassari, Italy Zhiying Tu, Harbin Institute of Technology, China Kay Tucci, Universidad de los Andes, Venezuela Bruno Tuffin, INRIA Bretagne Atlantique, France Alfonso Urquia, Universidad Nacional de Educación a Distancia, Spain Timo Vepsäläinen, Tampere University of Technology, Finland Maria Joao Viamonte, Instituto Superior de Engenharia do Porto, Portugal Manuel Villen-Altamirano, Universidad de Malaga, Spain Friederike Wall, Alpen-Adria-Universität Klagenfurt, Austria Hao Wang, Aalesund University College, Norway Frank Werner, Otto-von-Guericke-Universität Magdeburg, Germany Philip A. Wilsey, University of Cincinnati, United States Kuan Yew Wong, Universiti Teknologi Malaysia, Malaysia Li Xia, Tsinghua University, China Yiping Yao, National University of Defense Technology, China Gregory Zacharewicz, University of , France Marguerite Zarrillo, University of Massachusetts Darthmouth, United States František Zboril, Brno University of Technology, Faculty of information Technology, Czech Republic Durk Jouke van der Zee, University of Groningen, The Netherlands Houxiang Zhang, Aalesund University College, Norway Suiping Zhou, Middlesex University, UK Armin Zimmermann, Technische Universität Ilmenau, Germany Konstantinos Zografos, Lancaster University Management School, UK Borut Zupancic, University of Ljubljana, Faculty of Electrical Engineering, Slovenia xii Organization

Additional Reviewers

Eduardo Cesar, Universitat Autònoma de Barcelona, Spain Ruud van Damme, University of Twente, The Netherlands Christian Rohr, Brandenburg University of Technology Cottbus-Senftenberg, Germany Alexis Tcach, Universidad de Buenos Aired, Argentina Ning Yu, Georgia State University, United States

Invited Speakers

Eleni Karatza, Aristotle University of Thessaloniki, Greece Adelinde M. Uhrmacher, University of Rostock, Germany Richard Fujimoto, Georgia Institute of Technology, United States Pietro Terna, Università di Torino, Italy Contents

Numerical Assessment of Rubber Insulated Electric Cables Plants Efficiency Using Nitrogen and Steam Water as Curing Agents ...... 1 Gabriele Milani and Federico Milani Simulation Driven Policy Recommendations for Code Diversity ...... 21 Brady Tello, Michael Winterrose, George Baah and Michael Zhivich Multiple Runs in the Simulation of Stochastic Systems Can Improve the Estimates of Equilibrium Expectations ...... 35 Winfried Grassmann IoT Efficient Design Using WComp Framework and Discrete Event Modeling and Simulation ...... 49 S. Sehili, L. Capocchi, J.F. Santucci, S. Lavirotte and J.Y. Tigli Comparison of System Dynamics and Discrete Event Simulation Approaches ...... 69 Eylül Damla Gönül-Sezer and Zeynep Ocak Applying Petri Nets to Coalition Formation Modeling ...... 83 V. Mashkov, J. Barilla, P. Simr and J. Bicanek Virtualization Guided Tsunami and Storm Surge Simulations for Low Power Architectures ...... 99 Dominik Schoenwetter, Alexander Ditter, Bruno Kleinert, Arne Hendricks, Vadym Aizinger and Dietmar Fey High Precision Temperature Control for Injector Components of a Free-Electron Laser ...... 115 Kai Kruppa, Sven Pfeiffer, Gerwald Lichtenberg, Frank Brinker, Winfried Decking, Klaus Flöttmann, Olaf Krebs, Holger Schlarb and Siegfried Schreiber Agent-Based Wiki Page Simulation: A Conceptual Model...... 137 Roger W. McHaney and Jonathan Mossberg

xiii xiv Contents

Enabling Military Coalition Command and Control with Interoperating Simulations as a System of Systems...... 157 J. Mark Pullen

Author Index ...... 175 Numerical Assessment of Rubber Insulated Electric Cables Plants Efficiency Using Nitrogen and Steam Water as Curing Agents

Gabriele Milani and Federico Milani

Abstract The standard industrial process to produce medium voltage electric cables based on EPDM is characterized by a crosslinking obtained through peroxides with either nitrogen or steam (pressurized water vapor). Suboptimal material crosslinking is usually due to wrong vulcanization conditions, sometimes a consequence of a temperature drop of the vulcanizing agent along the production line. A Genetic Algorithm (GA) is here proposed to either predict the final crosslinking degree or select the optimal vulcanization conditions to maximize final insulator mechanical properties. In the first case, the algorithm is applied to a real steam water plant, where the final curing degree of the cable is experimentally obtained with Differential Scanning Calorimetry (DSC). The unexpected under-vulcanization of the item is a consequence of a steam temperature drop along the line, accurately predicted by the numerical approach proposed through the minimization of the difference between numerically predicted and experimentally determined crosslinking degree. In the second case, the GA is applied to a production line that uses nitrogen as vulcan- ization agent, with the aim of finding the optimal temperature and curing velocity to adopt in order to maximize final mechanical properties, such as tensile and tear strength. Both practical applications discussed show how production conditions could be automatically calculated according to the cable parameters, by increasing quality reliability and reducing scraps.

Keywords Power cables production lines Á EPDM elastomers Á Peroxide vulcanization Á Steam and nitrogen curing Á Genetic algorithm of optimization

G. Milani (&) Technical University of Milan, Piazza Leonardo da Vinci 32, 20133 Milan, Italy e-mail: [email protected] F. Milani CHEM.CO Consultant, Via J.F. Kennedy 2, 45030 Occhiobello, Rovigo, Italy

© Springer International Publishing Switzerland 2016 1 M.S. Obaidat et al. (eds.), Simulation and Modeling Methodologies, Technologies and Applications, Advances in Intelligent Systems and Computing 442, DOI 10.1007/978-3-319-31295-8_1 2 G. Milani and F. Milani

1 Introduction

The continuous vulcanization process of electric cables was developed more than fifty years ago. Nevertheless, the use of advanced mathematical studies to predict the insulator performance is not commonly applied, despite the fact that it could play an important role in the optimization of the final mechanical properties of the vulcanized items. In the last few years, high and medium voltage cables have been produced by using ethylene propylene co- and ter-polymers EPM/EPDM [1, 2], due to their good dielectric strength, aging resistance and partial discharge resistance. Another important aspect which leads to prefer the utilization of EPM/EPDM elastomers for electric cables is the possibility both to avoid the so called electrical treeing and to mix peroxides with EPM/EPDM rubber during the pre-processing phase at low temperatures. This allows, for instance, the use of a wide spread of commercial peroxides, as well as it ensures a well-defined level of vulcanization of the final product. In addition, it is worth noting that it is possible to obtain blends in particle form (pellets) where the peroxides are uniformly distributed and useable as plastic materials (e.g. PE). The use of various types of peroxide and co-agents to activate the crosslinking reaction [3] is preferred to the less expensive option of sulfur vulcanization [4], as it leads to well-defined levels of crosslinking of the final product. The level of crosslinking is strictly connected to the final properties of the material; for this reason, it represents a key parameter in cable production. The industrial process of cross-linking [5, 6] of power cables is usually obtained by means of horizontal (catenary continuous vulcanization CV) production lines, Fig. 1. A heating zone is followed by a cooling phase with water and/or air, as schematically depicted in Fig. 2. In the first phase, cross-linking of polymer is obtained by dry or wet curing under high pressure in a tube filled with nitrogen or steam water (less expensive). Process starts in an extruder (1), where the conductor (Al or Cu) is coated with the extruded EPM/EPDM. Then the cable enters the heating zone (2), filled with a high pressure vulcanizing agent. Finally, the cable is cooled to the ambient temperature in two steps: in the first step (3), cold water is used, finally the cable is leaved and cooled in the surrounding air by free convection (4). During the continuous process, in the curing tube the heat is transferred by convection and radiation [6, 7], see Fig. 2. Generally, the vulcanization tube length for CV lines can be up to 150 m (usually 60–80 m [8]). Two are the main problems that producers are called to tackle, depending if the curing process is done with nitrogen or steam water. For nitrogen vulcanization, which is more expensive and sensible to oxygen percentage under curing but provides much more uniformly cured items, it is interesting to numerically estimate the production parameters which, in general, may optimize rubber final mechanical properties. They are exposition time, temperature of the heated part of the tube filled with nitrogen and (less important) temperature and flow rate of the cooling water. Numerical Assessment of Rubber Insulated Electric Cables Plants … 3

Capstan Extruders

curing zone

from 5 to 15 m

pay off take up cooling zone

Capstan

<150 m

P: vapour pressure measure T1 evaluation

P extruder 2 m T1 =270 mm diameter change slope 11° slope 3° 1 =210 mm 3 total length

Heating tota phase l length of the line: 103 m Heating phase

Suspended cable for 30-35 meters, then resting on the lower part of the tube

T21 , if not submerged normally 10-20°C less than T

T3 and T 4 always submerged: from 70°C to 35°C

capstan

T2 T 5600 mm 3 T4

5 m 4 m 4 m Free surface water level Water pool Cooling phase

Fig. 1 Real industrial production line considered

From a practical point of view, exposition time is controlled by the production line speed, once that the vulcanization tube length is fixed. Usually, production plants have a vulcanization length which is difficultly changeable, being closely related to initial design specifications. Manufacturers select the aforementioned variables values on their own, usually following their experience on this field. Therefore, it appears particularly attractive from both a theoretical and practical point of view to purpose an optimization 4 G. Milani and F. Milani

Fig. 2 Schematic representation of heating phase (Phase I) and cooling phase with water (Phase II) approach to use instead of experience, in order to confirm and/or improve manu- facturers choices. Conversely, when dealing with steam water, one of the problems is the decrease in crosslinking temperature, which is experienced between the extruder head and the beginning of the steam pipe, due to the instability of the steam/water thermo- dynamic equilibrium. Typically, a barometer and a thermometer are placed at the beginning of the vulcanization line, not far from the extruder head, in order to monitor the steam conditions. Several plants are not equipped with a continuous and reliable temperature control along the line. This absence leads to poor control of the crosslinking conditions. In the case study analyzed by the authors, for instance, oscillations in the temperature at the end of the heating phase between 10 and 25 % of the proposed temperature were experienced. Such temperature drops along the line have meaningful consequences for the final crosslinking, especially for thick cables, where the core has a sensibly lower degree of crosslinking as it is subjected to a lower temperature. A Genetic Algorithm (GA) is here proposed to either predict the final crosslinking degree or select the optimal vulcanization conditions to maximize final insulator mechanical properties. In the first case, the algorithm is applied to a real steam water plant, where the final curing degree of the cable is experimentally obtained with Differential Scanning Calorimetry (DSC). The unexpected under-vulcanization of the item is a consequence of a steam temperature drop along the line, accurately predicted by the numerical approach proposed through the minimization of the difference between numerically predicted and experimentally determined crosslinking degree (fitness function). Each individual of the GA is represented by temperatures at different positions in the pipe. In the second case, the GA is applied to a production line that uses nitrogen as vulcanization agent, with the aim of finding the optimal temperature and curing velocity to adopt in order to maximize final mechanical properties, such as tensile and tear strength.