Induction Generator Based More Electric Architectures For

INDUCTION GENERATOR BASED MORE ELECTRIC ARCHITECTURES

FOR COMMERCIAL TRANSPORT AIRCRAFT

Yijiang Jia

APPROVED BY SUPERVISORY COMMITTEE:

Kaushik Rajashekara, Chair

Bilal Akin, Co-Chair

Babak Fahimi

Hoi Lee

Prasanna U R Copyright c 2016

Yijiang Jia

FOR COMMERCIAL TRANSPORT AIRCRAFT

YIJIANG JIA, BS, MS

DISSERTATION

Presented to the Faculty of

The University of Texas at Dallas

in Partial Fulﬁllment

of the Requirements

for the Degree of

DOCTOR OF PHILOSOPHY IN

ELECTRICAL ENGINEERING

THE UNIVERSITY OF TEXAS AT DALLAS

December 2016 ACKNOWLEDGMENTS

I would like to express my most sincere gratitude to my advisor, Dr. Kaushik Rajashekara, for his insightful guidance and patient support throughout my entire study toward the PhD. Dr. Rajashekara educated me with valuable critiques and constructive suggestions as a great academic advisor, and inspired me with warm-hearted kindness and enthusiastic encourage- ment as an admired mentor.

I would like to thank Dr. Akin, Dr. Fahimi, Dr. Lee, and Dr. Prasanna for being on my dissertation committee. I gratefully appreciate the time they spend on my behalf.

I would also like to thank Dr. Hao Huang from GE Aviation for his valuable advice and enlightenment for my research.

I would also like to thank all of my colleagues in the Power Electronics and Drives Laboratory for their helpful advice to my research and enjoyable time we spend together after work.

I would like to express my great appreciation to my wife, Yinghui, my parents, Donglin and Hong, for all their love to me.

October 2016

iv INDUCTION GENERATOR BASED MORE ELECTRIC ARCHITECTURES

FOR COMMERCIAL TRANSPORT AIRCRAFT

Yijiang Jia, PhD The University of Texas at Dallas, 2016

Supervising Professor: Kaushik Rajashekara, Chair

In the trend toward more electric aircraft, optimizing the performance of the new electrical power system in terms of reliability, fault-tolerance, size, weight, eﬃciency and cost is quite a challenging task, in which the type of the generator has great impact on the overall performance of the system. This dissertation explores and evaluates the option of using an induction generator for the distributed electrical power system of commercial transport more electric aircraft. In this dissertation, induction generator based electrical power generation and management system architectures are developed for both the main engine generation system and auxiliary power unit system. The application of induction generator in the proposed systems improves the system power density compared to synchronous generator based systems, and avoids the excessive faulty current issue caused by permanent magnet (PM) generators.

In the main engine generation system, an induction generator based AC/DC hybrid electric power generation system under twin-shaft twin-generator concept is proposed. The proposed AC/DC hybrid generation architecture supplies constant voltage variable frequency power directly from the generator winding terminals, and enables load sharing between the two engine shafts. Control schemes are developed to regulate the AC load voltage and coordinate DC power generation between the two generators. The feasibility of operation of the proposed

v system is demonstrated by both computer simulation and hardware-in-the-loop real-time emulation.

An auxiliary power unit (APU) that allows the regenerative power from the actuators to be absorbed by the turbine shaft of the APU is proposed. An open-end winding induction starter/generator is used to provide direct power flow path from the electro-hydrostatic actuators (EHAs) and/or electro-magnetic actuators (EMAs) to the power source, and to create a separate electric actuation bus without significant additional hardware requirement. A closed-loop control scheme for regulating both main DC bus and actuation DC bus voltages in aircraft emergency power mode is developed and verified by simulation in MATLAB/Simulink.

A modular back-up power link unit for re-configurable fault-tolerant actuation system architecture is also proposed to provide additional power supply path for the flight safety critical actuator loads in the proposed auxiliary power unit based regenerative power management strategy. A closed-loop control scheme for extracting constant and steady power flow from the primary power source through the modular back-up power link unit is developed and verified by simulation in MATLAB/Simulink.

The proposed more electric architectures in this dissertation provide solutions for electrification development of aircraft systems in terms of enhancing the electric power generation capacity of the aircraft, reducing the hardware requirement of the electric power generation and distribution system, managing the high peak and regenerative power flow from the EHA/EMAs, and enhancing the reliability and availability of the flight safety critical actuation system and the regenerative power management system.

vi TABLE OF CONTENTS

ACKNOWLEDGMENTS ...... iv ABSTRACT ...... v LIST OF FIGURES ...... x LIST OF TABLES ...... xiv CHAPTER 1 INTRODUCTION ...... 1 1.1 Motivation ...... 1 1.2 Research objectives ...... 3 1.2.1 Main engine electrical power generation system architecture and electric starter/generators ...... 3 1.2.2 Regenerative power management architecture and auxiliary power unit 4 1.2.3 Re-configurable fault-tolerant actuation system architecture using electro- hydrostatic or electro-mechanical actuators ...... 5 1.3 Summary of dissertation organization ...... 6 CHAPTER 2 CURRENT TRENDS AND CHALLENGES OF MORE ELECTRIC AIRCRAFT ARCHITECTURE ...... 8 2.1 Conventional and more electric aircraft architectures ...... 8 2.2 Trends of electrification and state-of-the-art architectures of the major subsystems of more electric aircraft ...... 12 2.2.1 More electric engine: load sharing and new starter/generator . . . . 13 2.2.2 Progressions in electrical power system ...... 19 2.2.3 Electrification of hydraulic systems: power management and load separation ...... 24 2.3 Future trends and expectations for the next generation more electric aircraft Architectures ...... 27 CHAPTER 3 INDUCTION GENERATOR BASED AC/DC HYBRID POWER GEN- ERATION SYSTEM FOR MORE ELECTRIC ENGINE ...... 32 3.1 Introduction: AC and DC primary main engine electric power generation architectures ...... 32 3.2 Direct AC power generation architectures for frequency insensitive loads on high pressure spool ...... 36 3.3 Proposed AC/DC hybrid power generation architecture ...... 38

vii 3.4 High pressure spool generation subsystem modeling and operating principle in generating mode: instantaneous power control theory based approach . . . 41 3.5 High pressure spool generation subsystem modeling and operating principle in generating mode: field orientation control theory based approach . . . . . 45 3.6 Low pressure spool generation subsystem operating principles and DC voltage regulation ...... 50 3.7 Closed-loop control scheme for generating mode of operation ...... 50 3.7.1 Instantaneous power control theory based control scheme for high pressure spool generation subsystem ...... 52 3.7.2 Field oriented control theory based control scheme for high pressure spool generation subsystem ...... 54 3.7.3 Control scheme for low pressure shaft generation subsystem . . . . . 56 3.8 Computer simulation results for instantaneous power control theory based control scheme ...... 57 3.9 Computer simulation results for field orientation control theory based control scheme ...... 60 3.10 Hardware-in-the-loop real-time emulation results ...... 64 3.11 Summary ...... 69 CHAPTER 4 AN INDUCTION GENERATOR BASED AUXILIARY POWER UNIT FOR POWER GENERATION AND MANAGEMENT SYSTEM FOR MORE ELEC- TRIC AIRCRAFT ...... 70 4.1 Introduction ...... 70 4.2 Open-end winding induction generator and inverter/rectirfier unit model . . 73 4.3 System operating principle ...... 76 4.3.1 Self-start mode of operation ...... 77 4.3.2 Main engine start mode of operation ...... 77 4.3.3 Cooling mode of operation ...... 78 4.3.4 Emergency power mode of operation ...... 79 4.4 System operating constraints and design considerations ...... 80 4.5 Control scheme in emergency power mode ...... 83 4.6 Simulation results ...... 84 4.7 Summary ...... 89

viii CHAPTER 5 A MODULAR BACK-UP POWER LINK UNIT FOR A RE-CONFIGURABLE FAULT-TOLERANT ACTUATION SYSTEM ARCHITECTURE WITH SEPARATED POWER SUPPLY IN MORE ELECTRIC AIRCRAFT ...... 90 5.1 Introduction ...... 91 5.2 A re-conﬁgurable fault-tolerant actuation system architecture with modular back-up power links ...... 94 5.3 Modeling and operation of the modular back-up power link unit ...... 98 5.4 Closed-loop control scheme of the modular back-up power link unit for emergency charging ...... 103 5.5 Simulation results ...... 104 5.6 Summary ...... 108 CHAPTER 6 CONCLUSION AND FUTURE WORK ...... 109 6.1 Conclusions ...... 109 6.2 Future work ...... 111 REFERENCES ...... 113 BIOGRAPHICAL SKETCH ...... 122 CURRICULUM VITAE

ix LIST OF FIGURES

2.1 Non-propulsive power distribution of conventional aircraft ...... 9 2.2 A possible non-propulsive power system concept of more electric aircraft . . . . 12 2.3 More electric engine system ...... 13 2.4 Diagram of load sharing between multiple spools ...... 14 2.5 The power optimized aircraft engine ...... 15 2.6 Diagram of main engine start using electrical power from APU ...... 17 2.7 Increasing electrical power demand in commercial air transport market . . . . . 19 2.8 IDG based constant voltage constant frequency electrical power system for conventional aircraft ...... 21 2.9 Constant voltage variable frequency electrical power system in Boeing 787 . . . 23 2.10 Potential DC primary electrical power system for more electric aircraft . . . . . 24 2.11 Actutators for aircraft flight control systems ...... 25 2.12 A potential electrical power system architecture with separated actuation DC bus 27 2.13 Schematic diagram of hybrid electric propulsion architecture ...... 28 2.14 Schematic diagram of hybrid turbo electric propulsion architecture ...... 29 3.1 System configuration of a wound field synchronous generator based AC primary generation system ...... 34 3.2 A potential system configuration of a DC primary generation system ...... 35 3.3 Circuit diagram of conventional shunt connected induction generation system . . 36 3.4 Inverter-load topology in the battery compensated series connected induction generation system ...... 37 3.5 System configuration of the proposed Open-end Winding Induction Generation System ...... 38 3.6 System configuration of the induction generator based AC/DC hybrid generation system ...... 39 3.7 Electrical system diagram of the induction generator based AC/DC hybrid generation system ...... 40 3.8 Starter mode of operation of the induction generator based AC/DC hybrid generation system ...... 41

x 3.9 Generator mode of operation of the induction generator based AC/DC hybrid generation system ...... 42 3.10 Operating constraints of the high pressure spool generation subsystem . . . . . 47 3.11 Contour maps of d-q axis current commands for AC load current and generator torque reference sweep test ...... 49 3.12 Overall closed-loop control scheme for the proposed AC/DC hybrid generation system ...... 51 3.13 Closed-loop current oriented control scheme for the HP spool generation subsystem 52 3.14 Basic configuration of conventional phase locked loop ...... 53 3.15 Closed-loop rotor flux oriented control scheme for the HP spool generation subsystem ...... 54 3.16 Closed-loop rotor flux oriented control scheme for the LP spool generation subsystem ...... 56 3.17 The DC link capacitor voltage characteristics of the proposed HP spool generation system controlled by current oriented control scheme ...... 57 3.18 The AC load voltage characteristics of the proposed HP spool generation subsystem controlled by current oriented control scheme ...... 58 3.19 The three phase current characteristics of the proposed HP spool generation subsystem controlled by current oriented control scheme ...... 58 3.20 The electromagnetic torque characteristics of the HP spool generator in proposed HP spool generation subsystem controlled by current oriented control scheme . . 59 3.21 The P and Q axis voltage characteristics of the HP spool generator in proposed HP spool generation subsystem controlled by current oriented control scheme . . 59 3.22 The P and Q axis current characteristics of the HP spool generator in proposed HP spool generation subsystem controlled by current oriented control scheme . . 60 3.23 The DC bus voltage regulation characteristics of the proposed AC/DC hybrid generation system ...... 61 3.24 The AC load voltage regulation characteristics of the proposed AC/DC hybrid generation system ...... 61 3.25 The electromagnetic torque characteristics of HP and LP generators in the proposed AC/DC hybrid generation system ...... 62 3.26 The d, q-axis current characteristics of the HP generator in the proposed AC/DC hybrid generation system ...... 62 3.27 The d, q-axis current characteristics of the LP generator in the proposed AC/DC hybrid generation system ...... 63

xi 3.28 Hardware-in-the-loop emulation implementation for the proposed AC/DC hybrid generation system ...... 65 3.29 The DC bus voltage regulation characteristics of the proposed AC/DC hybrid generation system ...... 66 3.30 The AC load voltage regulation characteristics of the proposed AC/DC hybrid generation system ...... 66 3.31 The phase A and B current characteristics of the HP generator in the proposed AC/DC hybrid generation system ...... 67 3.32 The phase A and B current characteristics of the LP generator in the proposed AC/DC hybrid generation system ...... 67 3.33 The electromagnetic torque characteristics of the HP generator in the proposed AC/DC hybrid generation system ...... 68 3.34 The electromagnetic torque characteristics of the LP generator in the proposed AC/DC hybrid generation system ...... 68 4.1 The need of direct power path between the aircraft electrical power source and the highly dynamic actuator loads ...... 71 4.2 The proposed power generation and management system architecture with regenerative power absorption capability ...... 72 4.3 The proposed power generation and management system architecture with regenerative power absorption capability ...... 74 4.4 Self-starting mode of operation for the proposed auxiliary power unit ...... 77 4.5 Main engine start mode of operation for the proposed auxiliary power unit . . . 78 4.6 Cooling mode of operation for the proposed auxiliary power unit ...... 79 4.7 Emergency power mode of operation for the proposed auxiliary power unit . . . 80 4.8 Unity power factor operation on main DC network side inverter/rectiﬁer unit . . 82 4.9 Closed-loop control scheme for the proposed APU for power generation and management system in emergency power mode ...... 84 4.10 The main DC bus voltage characteristics of the proposed APU for power generation and management system ...... 85 4.11 The electric actuation DC bus voltage characteristics of the proposed APU for power generation and management system ...... 86 4.12 The APU generator electromagnetic torque characteristics of the proposed APU for power generation and management system ...... 86

xii 4.13 The APU generator d and q axis current characteristics of the proposed APU for power generation and management system ...... 87 4.14 The main DC bus input power characteristics of the proposed APU for power generation and management system ...... 87 4.15 The electric actuation DC bus input power characteristics of the proposed APU for power generation and management system ...... 88 4.16 The APU generator output power characteristics of the proposed APU for power generation and management system ...... 88 5.1 Simplified block schematic diagram of a conventional hydraulic actuator with multiple redundancy ...... 91 5.2 Electrical backup hydraulic actuator used on Airbus A380 ...... 92 5.3 The demand of redundant power path for different type of electrical power system with separated actuator buses ...... 93 5.4 The proposed re-configurable fault-tolerant actuation system architecture with modular back-up power links for separated actuator loads in MEA ...... 95 5.5 Normal mode of operation of the proposed re-configurable fault-tolerant actuation system architecture ...... 96 5.6 Emergency charging mode of operation of the proposed re-configurable fault- tolerant actuation system architecture ...... 97 5.7 Normal mode of operation of the proposed modular back-up power link unit . . 98 5.8 Emergency charging mode of operation of the proposed modular back-up power link unit ...... 99 5.9 Equivalent circuit of the proposed modular back-up power link unit in emergency charging mode ...... 102 5.10 Closed-loop control scheme for the proposed modular back-up power link unit in emergency charging mode ...... 104 5.11 The d and q axis current characteristics of the proposed modular back-up power unit for re-configurable fault-tolerant actuation system architecture ...... 105 5.12 The d axis voltage characteristics of the proposed modular back-up power unit for re-configurable fault-tolerant actuation system architecture ...... 106 5.13 The q axis voltage characteristics of the proposed modular back-up power unit for re-configurable fault-tolerant actuation system architecture ...... 106 5.14 The three phase current characteristics of the proposed modular back-up power unit for re-configurable fault-tolerant actuation system architecture ...... 107 5.15 The emergency power battery voltage and current characteristics of the proposed modular back-up power unit for re-configurable fault-tolerant actuation system architecture ...... 107

xiii LIST OF TABLES

2.1 Squirrel cage induction generator vs. wound ﬁeld synchronous generator . . . . 18 2.2 NASA goals for future subsonic aircraft ...... 28

xiv CHAPTER 1

INTRODUCTION

1.1 Motivation

According to the statistics and the forecasts from the International Civil Aviation Organi- zation (ICAO), the revenue passenger-kilometres (RPK) of global air transport has grown at an average yearly rate of 6.2% since 2003, and will continue to grow at an average rate of

4.5% until the year 2030 [1]. The promising outlook of air transport growth comes with great commercial and environmental demands on aircraft manufacturers. The aircraft passengers and airline operators expect continuous improvements in terms of ﬂight safety, availability, services quality and operating cost. In the meantime, the ICAO requires the future aircraft operations to limit or reduce the aircraft noise and the impact of aviation emissions on local air quality and global climate [2]. As a result, in 2009, the International Air Transport As- sociation (IATA) and other major worldwide organizations in aviation industry announced the collective commitments that the whole industry will achieve: a 1.5% average annual improvement in fuel eﬃciency from 2009 to 2020; carbon neutral growth by 2020; and a

50% absolute reduction in carbon emissions by 2050 (compared to 2005 levels). In 2011, the

European Commission and Advisory Council for Aeronautics Research in Europe (ACARE) announced their aviation targets for 2050 are to: allow a 75% reduction in CO2 emissions per passenger kilometre and a 90% reduction in NOx emissions; and a 65% reduction of the perceived noise emission of ﬂying aircraft (relative to the capabilities of typical new aircraft in 2000) [3].

The above expectations and requirements have raised a trend of adopting the concept of more electric aircraft (MEA) architectures. The MEA concept aims the step-by-step removal or reduction of non-electrical secondary power usage (e.g., hydraulic power, pneumatic power and mechanical power) on aircraft, and invokes a new distributed electrical power system

1 to deliver higher quality non-propulsive power to the aircraft loads [4, 5]. The perceived benefits of adopting the MEA initiative include [6]: reductions of aircraft-empty weight, cost of ownership, specific fuel consumption (SFC), installation costs, maintenance costs and turn-around times; enhancements of aircraft range and system reliability. These benefits can potentially help the aircraft manufacturers to meet the market expectation and upcoming legislation requirement.

Over the last thirty years, remarkable progress has been achieved in the aircraft development moving toward MEA, where many hydraulic, mechanical, and pneumatic powered subsystems have been fully or partially replaced by electrical systems [7]. However, the magnitude of the benefits and negative consequences for adopting MEA initiative varies from different technical approaches which includes new electrical power system architectures and different generator types.

Optimizing the performance of the new electrical power system in terms of reliability, fault-tolerance, size, weight, efficiency and cost is quite a challenging task, in which the type of the generator has great impact on the overall performance of the system. Wound-field synchronous generator is kept as the main engine generator in the most recent commercial transport more electric aircraft (e.g., Boeing 787 and Airbus 380), whereas permanent magnet (PM) synchronous generator and switched reluctance (SR) generator are prevalent candidate in recently published relevant literature [8, 9, 10, 11, 12]. Using induction generator for MEA application is not much reported in the literature [13, 14, 15] due to the fact that the excited rotor cage makes induction generator naturally less compact and efficient compared to PM generator. However, induction generator is qualified as a fail-safe machine which immune the problem of excessive fault current in PM generator. Besides, induction generator based generation system presents higher overall power density compared to both wound-field synchronous generator and SR generator based system. In addition, the internal impedance of induction generator is the lowest among all type of generators.

2 This dissertation explores and evaluates the option of using induction generator for the distributed electrical power system of MEA. New induction generator based electrical power generation and management system architectures are developed for both the main engine generation system and auxiliary power unit system. A new actuation system architecture is also proposed to enhance the fault-tolerance of the power supply for the actuation system of the aircraft.

1.2 Research objectives

This dissertation focuses on three necessary development areas for transforming the conventional airplane toward the more electrical aircraft:

1. Main engine electrical power generation system architecture and electric starter/generators;

2. Regenerative power management system architecture and auxiliary power unit;

3. Re-conﬁgurable fault-tolerant actuation system architecture using electro-hydrostatic

or electro-mechanical actuators.

1.2.1 Main engine electrical power generation system architecture and electric

starter/generators

The focus of the work in the area of main engine electrical power generation system architecture and electric starter/generators (S/Gs) is to study the architectures for present and future MEA electrical power system (EPS) with diﬀerent S/G candidates. In the EPS of

MEA, both AC and DC electric power with multiple voltage levels are required for various aircraft loads. Neither the wound-ﬁeld synchronous generator (SG) based AC primary generation system in existing MEA [16] nor permanent magnet synchronous generator (PMG) based DC primary generation system in recent literature [17, 18, 19, 20] is able to meet all

3 the power requirements with optimized performance in terms of volume, weight, eﬃciency, reliability and cost.

An induction generator based AC/DC hybrid electric power generation system for MEA is proposed. Because of the self-excited capability of induction generator, the proposed AC/DC hybrid generation architecture can supply constant voltage variable frequency (CVVF) power directly from one side of the generator winding terminals like an AC primary generation system. On the other side of the generator winding terminals, the more advanced paralleled

DC bus power generation architecture from DC primary generation system is also reserved.

AC load voltage and coordinate DC power generation between the two generators. Both AC and DC output voltages of the system can be well-regulated with generator speed, AC and

DC side load and DC power output command variation. The feasibility of operation of the proposed system is demonstrated by both computer simulation and hardware-in-the-loop

(HIL) real-time emulation.

1.2.2 Regenerative power management architecture and auxiliary power unit

The focus of the work in the area of regenerative power management system architecture and auxiliary power unit (APU) is to investigate the issues caused by hydraulic system electriﬁcation and search for a solution of managing the high peak and regenerative power

ﬂow from the electro-hydrostatic actuators (EHAs) or electro-mechanical actuators (EMAs) without signiﬁcant additional hardware requirement. Dissipating the regenerated energy with associated cooling devices or enabling energy recovery by adding electrical energy storage elements (ESEs) such as ultra-capacitors and batteries requires considerable additional hardware installment [21].

4 An APU for power generation and management system that allows the regenerative power from the actuators to be absorbed by the turbine shaft of the APU is proposed. An open- end winding induction starter/generator is used to provide direct power flow path from the EHA/EMAs to the power source, and to create a separate electric actuation bus without significant additional hardware requirement. In this way, the operation of aircraft main DC power network is prevented from perturbation and disturbance caused by the actuators. The configuration and modeling of the proposed induction generator based APU for power generation and management system are presented. System operation for all operating conditions of the aircraft is analyzed. A closed-loop control scheme for regulating both main DC bus and actuation DC bus voltages in aircraft emergency power mode is developed based on FOC and instantaneous power theory. The feasibility of operation of the proposed system is demonstrated by simulation in MATLAB/Simulink.

1.2.3 Re-conﬁgurable fault-tolerant actuation system architecture using electro- hydrostatic or electro-mechanical actuators

The focus of the work in the area of re-configurable fault-tolerant actuation system architecture using electro-hydrostatic or electro-mechanical actuators is to explore the options for reconfigurability of the separated electrical actuation system. Separate electric actuator bus can protect the main electrical power network from disturbance caused by the EHA/EMAs [22]. As one of the most critical loads on board, the power supply for actuator bus requires superior fault-tolerance. Aircraft emergency batteries are usually used to prevent a complete power loss in case of equipment failure occurs in the primary power path for the separated actuator bus. However, if an equipment failure occurs in the primary power link (such as the new APU mentioned in the previous subsection) between aircraft main power grid and separated actuation power bus, the flight safety critical actuator loads are isolated from the rest of the aircraft electrical power network. A back-up power path can reduce the requirement of the aircraft emergency batteries and improve the system reliability and availability.

5 A modular back-up power link unit for re-configurable fault-tolerant actuation system architecture is proposed. Each individual actuator motor drive is configured as three-phase H-bridge converter in normal operation. In case of the failure occurs in the path towards primary power source of the actuation bus, the three-phase H-bridge converter can be transformed into two six-switch two level inverter/rectifier. One of the inverter/rectifier is connected with the electrical bus of primary power source, whereas the separated actuator bus is connected with the other inverter/rectifier. In this way, each of the actuator motor drives can act as a back-up power link between the main source and the separated bus. Modeling and operation of the proposed modular back-up power link units are discussed. A closed- loop control scheme for extracting constant and steady power flow from the primary power source through the modular back-up power link unit is developed. The feasibility of the operation of proposed modular back-up power link unit is demonstrated by simulation in MATLAB/Simulink.

1.3 Summary of dissertation organization

This dissertation is organized as follows: Chapter 2 provides a detailed review of previous work relevant to this study. The defini- tive directions and research gaps are identified for the proposed work. The electrification trends and technology advancements of the main engine, pneumatic and hydraulic systems in MEA are discussed. The state-of-the-art architectures of the major subsystems and equipment of MEA including electrical power system, main engine starter/generator, auxiliary power unit and flight control actuation system are presented. The challenges to make head- way for next generation MEA are examined. Chapter 3 presents an induction generator based AC/DC hybrid generation system for MEA. The proposed more electric engine based system provides CVVF power generation directly from the generator terminals without external excitation. The configuration of

6 AC/DC hybrid generation is presented. Analysis based on instantaneous power control and field oriented control theory is carried out to explain the proposed twin-spool twin-generator AC/DC hybrid generation method. Closed-loop control schemes for AC and DC voltage regulation of the proposed system are developed based on FOC and instantaneous power control theory. The feasibility of operation of the proposed system is demonstrated by both computer simulation and hardware-in-the-loop real time emulation. Chapter 4 presents an auxiliary power unit (APU) for power generation and management system to supply/absorb the highly dynamic power demand/regeneration from the EHA/EMAs. The configuration and modeling of the proposed induction generator based APU for power generation and management system are presented. System operation for all operating conditions of the aircraft is analyzed. A closed-loop control scheme for regulating both main DC bus and actuation DC bus voltages in aircraft emergency power mode is developed based on FOC and instantaneous power theory. The feasibility of operation of the proposed system is demonstrated by simulation in MATLAB/Simulink. Chapter 5 presents a modular back-up power link unit for re-configurable fault-tolerant actuation system architecture with separated power supply in more electric aircraft. The configuration and operating principle of the proposed re-configurable fault-tolerant actuation system are presented. Modeling and operation of the proposed modular back-up power link units are discussed. A closed-loop control scheme for extracting constant and steady power flow from the primary power source through the modular back-up power link unit is developed. The feasibility of the operation of proposed modular back-up power link unit is demonstrated by simulation in MATLAB/Simulink. Chapter 6 summarizes the key conclusions and contributions of the work and proposes recommendations for the follow-on research in the areas of main engine electrical power generation and internal starter/generator; regenerative power management and auxiliary power unit; and fault-tolerant operation of MEA actuation system.

7 CHAPTER 2

CURRENT TRENDS AND CHALLENGES OF MORE ELECTRIC

AIRCRAFT ARCHITECTURE

Over the past forty years, considerable amount of research has made remarkable progress to move towards more electric aircraft systems. In recent MEA systems, numerous mechanically, hydraulically, or pneumatically powered subsystems are partially or even completely replaced by electrical counterparts. The architectures of the entire aircraft system and corresponding subsystems are undergoing revolutionary changes and facing immense challenges.

This chapter reviews the trends of technology advancements toward MEA and the state-of- the-art architectures in MEA (sub)systems. The passage to future MEA architecture and the challenges along the electriﬁcation path are also discussed.

2.1 Conventional and more electric aircraft architectures

A basic schematic of the non-propulsive power distribution of conventional aircraft is shown in Figure 2.1 [23]. In conventional aircraft, the jet fuel is consumed by the gas turbine engine to generate power. Most of this power is used for propulsion of the aircraft, whereas the remainder is converted into pneumatic, hydraulic, mechanical and electrical power, where:

• Pneumatic power is obtained from the high pressure compressors of the gas turbine en-

gine. This form of power is used to supply high pressure bleed air for the environmental

control system (ECS) and wing anti-icing system. For a 300 passenger civil aircraft

with 40 MW equivalent propulsion thrust, the pneumatic power needed is about 1.2

MW [24].

• Hydraulic power is provided by a central hydraulic pump. The actuation systems for

primary and secondary ﬂight control; landing gear deployment, retraction and braking;

8 Primary Primary controls controls

Commercial Propulsion thrust Secondary loads controls APU (equivalent) ≈ 40 MW Mechanical power ≈ 100 kW Electrical Pneumatic power ≈ 1.2 MW distribution Hydraulic power ≈ 240 kW Landing Central gear hydraulics Electrical power ≈ 200 kW Engine Generator systems Environmental control Ice Gearbox protection Engine

Figure 2.1. Non-propulsive power distribution of conventional aircraft [23]

engine actuation and other ancillary systems are powered by hydraulic power. For a

300 passenger civil aircraft with 40 MW equivalent propulsion thrust, the pneumatic

power needed is about 240 kW [24].

• Mechanical power is obtained from the gas turbine engine shaft through mechanical

gearboxes. The form of power is used to drive various types of pumps and loads

including the central hydraulic pump, engine oil pumps, etc. For a 300 passenger civil

aircraft with 40 MW equivalent propulsion thrust, the mechanical power needed is

about 100 kW [24].

• Electrical power is generated by main engine generator which connected to the gas tur-

bine engine shaft through a variable ratio gearbox. Avionics, aircraft and cabin lighting,

galleys and other commercial loads are the main consumers of electrical power. For

9 a 300 passenger civil aircraft with 40 MW equivalent propulsion thrust, the electrical power needed is about 200 kW [24].

The performance of the hybrid non-propulsive power system of aircraft has been improved over the years, yet its complexity has grown even faster. Presently, the conventional non- propulsive power system still represents a major factor in aircraft malfunctions and failures. Furthermore, because of the inflexible infrastructure of the pneumatic and hydraulic systems, the leakage of the systems is generally difficult to locate or accessed, resulting in a great number of aircraft maintenance “down-times” and even flight delays. Optimizing the hybrid non-propulsive power system with incremental changes to existing products individually has become increasingly difficult. As a result, in late 1970’s, building on the advances in electric motor/generators and power electronic devices, the concept of all electric aircraft (AEA) appeared to offer many advantages [25]. In [26], the AEA concept was described as that all non-propulsive power take-off from the aircraft are electrical in nature, where the need for on-engine hydraulic power generation and bleed air take-off are completely removed. In early 1980’s, the National Aeronautics and Space Administration (NASA) sponsored the integrated digital/electrical aircraft (IDEA) studies which employs all electric secondary power systems and advanced digital flight control systems [26]. The technology advancement through early to mid 1980’s indicated that great benefits are to be gained from adopting a single-source electric power system which consists of no pneumatic and hydraulic secondary systems. However, for aerospace industry, the risks in- volved in this aircraft level comprehensive change in equipment technology was too large compared to the potential benefits. Alternatively, the industry chose a moderate approach to embrace the trend of electrification on aircraft. In late 1980’s, the United States (US) Air Force initiated an extensive research, development and demonstration program named more electric aircraft(MEA), aimed for collectively establish the capability of using electrical power to dramatically reduce or completely replace the centralized hydraulics on aircraft.

10 The MEA approach was rapidly accepted as the essential path to provide a more reliable, maintainable and battle-damage re-conﬁgurable military aircraft with less weight, volume and operation/support cost. Soon thereafter, the MEA concept began a global pursuit toward the electrically powered aviation power systems in the commercial aircraft community.

Several programs from worldwide have been started and signiﬁcantly impacted the research in this ﬁeld. The Totally Integrated More Electric Systems (TIMES) program sponsored by the United Kingdom (UK) Department of Trade and Industry (DTI) aimed to promote and utilize previously developed technologies into MEA within a collaborative working environment involving industrial companies, universities and academic associations [6]. The US Air

Force, Army, Navy and Marine scientists and engineers jointly devoted to the More Electric

Initiative (MEI) for improving the electrical power system reliability and system level power density of military aircraft. In Europe, the Power Optimized Aircraft (POA) project focused on optimize the management of electrical power in order to reduce the aircraft non-propulsive power usage and fuel consumption[27].

Through the work of researchers and engineers all over the world, the list of mechanical, pneumatic and hydraulic loads to be replaced by electrical counterparts are growing pro- gressively over time. Presently, novel means of generating, distributing, storing and using non-propulsive power on board are investigated and examined at aircraft level. A possible non-propulsive power system concept of MEA is shown in Figure 2.2 [23]. In this system, electric motors are installed to replace the mechanically driven fuel pump, ground air compressor started turbine engine, high stage bleed air powered cabin air compressor and hydraulic powered actuators. Electric systems are used for air conditioning, wing anti-icing system, and powering galley loads. For a 300 passenger civil aircraft with 40 MW equivalent propulsion thrust, the required electrical power is more than 1 MW. It is believed that the new electrical systems oﬀer far more options for future improvement than conventional systems in terms of reducing operating costs, fuel consumption and environmental impact

11 No gearbox Reduced engine bleed Primary Primary Local hydraulic source controls controls More electrical power

Cabin expansion More/full electrical & less mechanical Secondary generator Fuel pumping controls Engine Ancillaries Commercial loads More/full electrical & less pneumatic Main engine starting Cabin pressurization Air conditioning Landing Electrical Icing protection gear distribution More/full electrical & less hydraulic Engine Surface actuation systems Starter/ generator Landing gear Environmental Braking + Doors control Ice protection Electrical system power ≈ 1MW

Engine

Figure 2.2. A possible non-propulsive power system concept of more electric aircraft [23] from air travel. The reduction or removal of the bleed air usage in pneumatic system can signiﬁcantly improve the eﬃciency of the gas turbine engine; the reduction or removal of the mechanical and hydraulic systems can potentially lead to aircraft level weight reduction.

Furthermore, the electrical systems are much more easier to be diagnosed, controlled, and maintained compared to the conventional systems and presents superior re-conﬁgurability, which can help to improve the aircraft availability and cut down the scheduled maintenance eﬀort as well as the occurrence of unscheduled maintenance.

2.2 Trends of electriﬁcation and state-of-the-art architectures of the major sub-

systems of more electric aircraft

In the complicated aircraft system, the electriﬁcation of engine system and hydraulic system, as well as the improvement of electrical power system capacity, reliability, and power quality are matters of the utmost importance in steps toward MEA. The trends of electriﬁcation

12 HP Spool Turbine System Gen. Converter

LP Spool Motor/ Turbine System gen.

28 V Hydro. Oil pump Fuel pump pump Converter/ motor motor motor inverter

Controllers Inverter Inverter Inverter

Figure 2.3. More electric engine system [8] and state-of-the-art architectures of the above subsystems are investigated and presented in the following subsections.

2.2.1 More electric engine: load sharing and new starter/generator

The interest in the concept of MEA resulted in the development of new aircraft jet engine leading to the concept of more electric engine (MEE) by NASA in 1970’s [28]. As is shown in Figure 2.3 [8], the MEE initiative aimed to replace mechanically-driven engine accessories and oil, fuel and hydraulic pumps with electrical counterparts. The removal of mechanical drives and current pneumatic, hydraulic and lubrication systems can potentially reduce the weight, increase the eﬃciency and reliability of the engine. An aircraft gas turbine engine generally has more than one shaft, called high pressure (HP) shaft (or spool) and low pressure (LP) shaft (or spool). Some engines may have an

13 Combustor

HP Spool HP HP HPG Compressor Turbine

3 + AC/DC -

LP LP Spool LP LPG Compressor Turbine

3 AC/DC + - Inlet Nozzle

Figure 2.4. Diagram of load sharing between multiple spools [7] intermediate pressure (IP) shaft (or spool). In traditional aircraft systems, the electrical generators are driven by the HP shaft of the aircraft engine. The effect of electrical power take-off can sometimes have significant negative impact on the dynamics and control of the aircraft engine. For instance, during the transition from cruise to descent phase, the aircraft engine power is transiently reduced while maintaining high electrical power demand. This transition creates a possibility of engine instability and may require substantial electric load shedding. Furthermore, with the increasing electric power consumption in MEA, the above effect will be more severe if the electric power is solely extracted from the HP shaft of the gas turbine engine [29]. This issue has led the exploration of power take-off from multiple spools. As shown in Figure 2.4 [7], an extra electric generator can be coupled to the LP shaft of the engine. This additional means of power extraction enables the power sharing between the HP and LP spools [29, 30, 31]. In this way, the power generated from the LP spool could compensate for the decreased power from the HP spool so that the electrical power demand is not compromised. In 2008, the power optimized aircraft (POA) project [27], funded by European Commis- sion (EC), successfully tested a more electric engine (MEE) in both starting and generating

14 Figure 2.5. The power optimized aircraft engine [27] modes. The POA engine (shown in Figure 2.5 [27]) removed the external gearbox and used electrically-driven fuel pumps and vane actuation system. A permanent magnet embedded motor/generator is directly installed on the HP spool, and a switched reluctance generator connected to the LP engine spool through a 1:4.5 ratio gearbox. The POA project proved that the electrically-driven engine accessories are capable of replacing the conventional mechanically-driven systems. However, the overall performance of the POA engine in terms of aircraft total equipment weight, fuel economy, production costs and system reliability did not meet the initial expectation.

In MEE philosophy, the main engine start system is no longer dependent on pneumatic power. In B787, the air ducts and the air turbine starter (ATS) are eliminated. The main engine generators are used to be operated as electric motors to accelerate the HP spool of the main engine. Once the HP shaft reaches its self-sustaining speed, the combustor of the

15 engine will ignite and begin to work. In this way, the main engine generator connected to the HP spool becomes a starter/generator. The electric power needed for main engine start is provided by the APU generator(s). The electrical engine start process using power from

APU is illustrated in Figure 2.6.

In conventional aircraft electrical power generation system, wound ﬁeld synchronous machine are widely used as the main engine generator. However, using other type of electric machine for starter/generator (or generator) in main engine and APU generation system has been frequently evaluated and tested by many researchers and engineers [32, 33, 34]. Wound

ﬁeld synchronous machine (WFSM), permanent magnet synchronous machine (PMSM), switched reluctance machine (SRM) and induction machine (IM) are the popular candidates in this area [8]. The evaluation checklist of the potential candidate usually includes the power density of the machine and corresponding power generation system, machine fail- safety and fault-tolerance, robustness, eﬃciency, etc.

The dominance of wound field synchronous generators (WFSGs) in airborne applications can be mainly attributed to their simple and full control over its field flux, robust structure and decades of incremental improvements in system integration to the air-frame [35].

The advantages of the simple voltage control strategy of WFSG is maximized in both the constant frequency (CF) and variable frequency (VF) AC primary generation systems. In such systems, the WFSG can feed the AC loads directly from the generator terminals. The

AC output voltage of the WFSG can be regulated by controlling the field current through the brushless exciter. This exciter consists of a small power permanent magnet (PM) machine and a diode rectifier mounted on the generator shaft. The field flux of WFSG can be disabled in case of generator failure, which make it a fail-safe generator. However, the complex rotor structure makes the torque to inertia ratio of SG lower than other type of electric machines [8]. Moreover, the rotating diode bridge structure limits the top speed of the generator shaft. If the synchronous machine is used as a starter/generator, separate field

16 Main APU Engine Turbine Turbine

S/G S/G

3 3 Aircraft AC/DC AC/DC Control

Figure 2.6. Diagram of main engine start using electrical power from APU and armature voltage controls are required during its motoring operation. As a result, in the DC primary power generation system which proposed with the MEE concept for MEA, WFSG is no longer a favorable choice. Permanent magnet (PM) synchronous starter/generators become a popular candidate in the DC primary power generation system under MEE philosophy [9, 11]. PM S/Gs presents high power factor and high eﬃciency among all types of generators due to the loss-free magnet excitation. However, PM generators are not fail-safe machines and can produce excessive fault current in case of winding short circuit fault. Although multi-phase fault-tolerant PM generators have been investigated to limit the short-circuit fault current [36, 37, 38, 39], using PM generator as the main engine generator in aerospace application is still prohibited by Federal Aviation Administration (FAA). Furthermore, installing the PM generator close to the gas turbine engine may greatly aﬀect the system reliability since most permanent magnet materials are vulnerable to demagnetization under very high temperature. The SRM received considerable interest for its potential in integral S/G application on aircraft engines since late 1980’s [40]. The major advantages of SRM in this application include inherent fault tolerance due to its magnetic and electric isolated phases; high reliability because of the absence of permanent magnets; wide speed operation range and ability

17 Table 2.1. Squirrel cage induction generator vs. wound ﬁeld synchronous generator

Generator size Large Medium Generator type Cage IG DC-excited SG Cage IG DC-excited SG Power Rating (kW) 6000 6000 75 75 Speed (rpm) 60 60 1500 1500 Eﬃciency (%) 95.1 92 93.5 88 Inertia (kg m2) 30000 60000 0.75 0.673 Length (cm) 1500 1500 86.3 102.2 Width (cm) 410 473 45 39 Weight (kg) - - 385 480 to withstand very high temperature. These merits make SRM a suitable candidate as a direct-drive main engine S/G, which allows the elimination of gearbox and hydraulic accessories in conventional jet engine. The SR S/G has proved itself as a viable option from a series of designs and test results [40, 41]. However, this type of machine suﬀers from extremely high torque pulsations. Further, in order to mitigate the over-sized drive system

VA rating, SRM are usually designed to operate in deep saturation region, which results in insuﬃcient overload margin [42].

Squirrel-cage IMs are well known by their simplicity and ruggedness. Using IM as a generator can also beneﬁt from its lowest internal impedance among all other types of machines. Induction S/G has superior transient overload capability, and can also withstand harsh operating environment such as high temperature and vibration. Although due to the distributed stator winding, IM is not an inherently fault tolerant machine, the absence of permanent magnet still make IM a fail-safe generator candidate. A comparison between cage IG and WFSG in medium and large power level application are shown in Table 2.1.

Induction S/Gs present competitive performance compared to WFSGs, but do not share the safety concern of PMGs. Using squirrel-cage IM as S/G in airborne application was investigated by a series of research projects at University of Wisconsin in Madison in late 1980’s

[43, 44]. The research focus, however, was on the control of the entire drive system with a

18 Figure 2.7. Increasing electrical power demand in commercial air transport market [8, 39] high frequency resonant link. The possibility of using induction S/G in MEA architectures is still to be explored.

2.2.2 Progressions in electrical power system

For the past several decades, the aircraft electrical power system (EPS) has become increasingly complex. As is shown in Figure 2.7 [8, 39], the electrical power demand rises rapidly with continuously growing inventory of electrical devices on board. Environmental control system (ECS), wing anti-icing system, and actuation system for ﬂight control surfaces and landing gears are the main contributors of the growing electrical power demand. In the most recent MEA (e.g., B787), the total electrical power demand under cruising condition is exceeding 1 MW [16].

19 The primary function of the EPS is to generate, regulate and distribute electrical power throughout the aircraft. In large commercial transport aircraft, each gas turbine engine equips at least one generator for electric power generation. In order to meet the redundancy and extended range twin operations (ETOPS) requirements, back-up generators and additional power source such as auxiliary power unit (APU) are installed on board. The APU is a relatively small gas turbine engine which usually used to supply pneumatic, mechanical and electric power while the aircraft is on the ground, but it can also be used to provide auxiliary power in case of in-ﬂight engine failure. However, because of the reduced inlet air density at very high altitude, the power capacity of APUs is limited in some aircraft cruise condition [45]. In addition, many aircraft carry a ram air turbine (RAT) which can act as a small wind turbine and provide emergency power to the electric and hydraulic system on board. The RAT is generally deployed when all the engines and APU of the aircraft are failed. The power generated by the RAT is used to support the critical systems such as avionics in cockpit, ﬂight control system, emergency lighting for aisles and landing gears to operate long enough so that the aircraft can land safely. Various batteries are also installed in the aircraft to start the APU and provide extra back-up electric power when needed [46].

In most conventional commercial aircraft, the EPS uses a 115 V line to neutral AC voltage at 400 Hz line frequency. A 28 V DC bus is also equipped, and obtained by rectifying the 115

V 400 Hz AC through the transformer rectiﬁer units (TRUs) [47]. The primary electric power is usually generated by the electric generator(s) driven by the main gas turbine engine(s).

During the entire ﬂight mission, the engine shaft speed varies in a wide range. A few diﬀerent approaches are used to provide variable speed constant frequency power generation. The most commonly used method is the integrated drive generator (IDG). In an IDG, a constant speed drive (CSD) is deployed to work as an automatic gearbox maintaining a roughly constant generator shaft speed throughout the entire engine shaft speed variation range.

The generator output frequency can be maintained at 400 Hz with a tolerance less than

20 Engine Engine Turbine APU Turbine Turbine RAT

CSD CSD S/G G

S/G 3 3 S/G IDG IDG 3 3

115V 400Hz AC 115V 400Hz AC Distribution Distribution

TRU TRU

115V 400Hz 115V 400Hz 28V DC Bus AC Loads AC Loads

Figure 2.8. IDG based constant voltage constant frequency electrical power system for conventional aircraft [48]

10 Hz for a 2:1 ratio in engine speed variation. An IDG based constant voltage constant frequency EPS for conventional aircraft is shown in Figure 2.8 [48]. In all aircraft electrical power systems, many switches/circuit breakers are installed in order to disconnect the failed generators, loads, and buses from the rest of the system. The primary and secondary power distribution are monitored, controlled and protected by the integrated control centers (ICCs).

As a hydro-mechanical device, the CSD is extremely bulky and susceptible to be worn out.

Regular maintenance or replacement of the CSD is required with predeﬁned time-line in order to prevent potential failures [49].

In conventional aircraft, constant frequency electric power can also be generated by variable speed generators. Back-to-back converters or cycloconverters are used to convert the variable frequency AC output power from the generators to constant frequency at 400 Hz,

115 V AC power [50]. In B737, MD-90 and B777 airplanes, the back-to-back converter is

21 used to provide constant frequency power. In such system, the variable frequency voltage from the generator is ﬁrst rectiﬁed to DC power using an AC-DC converter, and converted back to three-phase AC power at 400 Hz, 115 V. The cycloconverters are mainly used in US military aircraft (e.g., F-18, U-2 and F-117, etc.). This technology requires a large number of power switches and complex control system, as a result, it has not been applied to any civil aircraft system [51].

In most recent MEA such as Boeing 787 (B787), Airbus A380, and A350 XWB, the conventional constant voltage constant frequency (CVCF) electric power system is replaced by constant voltage variable frequency (CVVF) system where the electric power frequency in the system changes proportional to the engine speed. The operating voltages of the VF systems can be either 115 or 230 VAC, while the operating frequency range varies from different engine and aircraft [52]. As shown in Figure 2.9 [16], a 230 VAC and 360 - 800 Hz CVVF electrical power system is used in Boeing 787. In this system, 230 VAC frequency insensitive loads are directly connected to the main AC distribution feeders. 115 VAC CVVF loads are supplied through auto transformer units (ATUs). Compared to conventional aircraft system, the total electrical power consumption of B787 increased dramatically. This significant increase of electrical power rating requires higher system voltage to mitigate the rising of feeder cable current. The potential growth in cable weight, power losses, and thermal management requirements may also be alleviated [53, 17]. The wide variation of the electric power frequency requires additional power electronic converters for the frequency sensitive aircraft loads such as the environmental control system (ECS) compressor motors and fans, electrically driven hydraulic pumps, Nitrogen generating systems (NGS), etc. Although the added converters increase the system cost and complexity, the freedom of operating the motors at variable speeds allows the motors to be run at higher speed and work at their optimum operating point along the load variation. Potential weight and volume reduction and performance improvement of the motors make significant positive trade-offs to the new

22 Engine APU Engine RAT Turbine Turbine Turbine

S/G S/G G S/G

3 3 3 3

230V 350-800Hz 230V 350-800Hz AC Distribution AC Distribution

ATU AC/DC TRU TRU AC/DC ATU

2 2 230V CVVF 115V CVVF +/-270V 28V DC Bus +/-270V 115V CVVF 230V CVVF AC Loads AC Loads DC Loads DC Loads AC Loads AC Loads

Figure 2.9. Constant voltage variable frequency electrical power system in Boeing 787 [16] system [12]. In addition, the application of CVVF electrical power system completely elim- inates the IDG or other variable speed constant frequency (VSCF) conversion devices from conventional aircraft. The removal of IDG improved the dispatch reliability of the MEA and reduced the maintenance effort requirement. The potential reduction of scheduled maintenance time and unexpected “grounded” time of the aircraft leads to significant savings for the airline operators [54, 55]. The concept of generation from multiple shafts and load sharing between the generators requires paralleling generators running at different fundamental frequencies [10]. In order to parallel these generators with enhanced efficiency and reduced size and weight, a DC primary generation system with power electronic converters is preferred as an advanced more electric architecture [56, 18, 19, 20]. As shown in Figure 2.10 [12], a high speed starter/generator and a low speed generator are connected to the HP and LP spool of the engine, respectively. In the engine starting process, the starter/generator on HP spool can operate as a motor to start the engine using the electrical power from the APU or ground power supply. In the flight mission, the power generated from the two generators are rectified and transmitted to

23 Engine Engine APU Engine Engine RAT LP HP Turbine HP LP Spool Spool Spool Spool

G S/G S/G G S/G G

AC/DC AC/DC AC/DC AC/DC AC/DC AC/DC 2 2 2 2 2 2

+/-270V DC Bus +/-270V DC Bus

DC/AC DC/AC TRU TRU DC/AC DC/AC

230V CVVF 115V CVVF +/-270V 28V DC Bus +/-270V 115V CVVF 230V CVVF AC Loads AC Loads DC Loads DC Loads AC Loads AC Loads

Figure 2.10. Potential DC primary electrical power system for more electric aircraft [12] a +/-270 V DC power bus. In this system, the +/-270 V DC loads can be directly connected to the main DC bus. However, the AC power demanded by the frequency insensitive AC loads (e.g., wing de-icing system, galleys, etc.) is ﬁrst converted to DC power by the active rectiﬁer of the generator, and inverted back to AC power through dedicated inverters. Such two-stage AC-DC-AC conversion adds extra losses and additional hardware to the system. In Boeing 787, power consumption of frequency insensitive AC loads (e.g., wing de-icing system, galleys, etc.) under cursing condition is close to 50% [57].

2.2.3 Electriﬁcation of hydraulic systems: power management and load separation

In conventional aircraft, the aircraft primary and secondary ﬂight control surfaces, braking, landing gears and other actuation functions are implemented by fault tolerant centralized redundant hydraulic systems. These hydraulic systems are operated by mechanically driven actuators, which is shown in Figure 2.11 (a) [58]. The current trend of electriﬁcation of

24 (a)

(b) (c)

Figure 2.11. Actutators for aircraft ﬂight control systems: (a)conventional mechanically driven servo-control actuator [58] (b) electro-hydrostatic actuator [59, 60, 61] (c) electro- mechanical actuator [62] the hydraulic systems is to replace the conventional servo-control actuators with electro- hydrostatic actuators (EHAs) or electro-mechanical actuators (EMAs) [59, 60, 61].

As shown in Figure 2.11 (b) [58], the EHA uses a electric motor and dedicated inverter to drive the hydraulic pump and provide actuation function. In most recent MEA such as

Airbus A380, EHAs are used as back-up actuators where a centralized hydraulic systems are still serving the ﬂight control system in normal operation. However, the structure of EHA allows the centralized hydraulic system including the hydraulic source and piping system to be eliminated. A de-centralized actuation system oﬀers reduced weight, volume, production and maintenance costs with the same level of safety performance [63].

25 The configuration of an EMA is shown in Figure 2.11 (c) [62]. EMAs do not require hydraulic fluid and reservoir, instead, a ball screw with gearbox is used to convert rotary motion from the electric motor into linear motion. EMAs are generally more efficient than EHAs, however, potential mechanical jamming make EMAs not viable for safety critical applications such as primary flight controls and landing gear deployment [64]. Although they have been already used in aeronautics, the application on aircraft is still limited to secondary flight controls or military aircraft [62]. The application of EHAs or EMAs brings new issues to the aircraft electrical power system. These actuators have highly dynamic power profile, which includes high peak power demand and power regeneration [65, 66, 67]. The issues could lead to EPS component size and weight penalties, potential thermal management challenges, and difficulty of fulfilling the bus power quality requirement from aircraft electric power characteristics standard (e.g., MIL-STD-704) [68, 69]. The most widely used power management method for EHA/EMAs on aircraft is to employ resistor banks paralleled with the actuators, dissipating the regenerated energy with associated cooling devices. This solution renders the weight, volume, and efficiency of the overall system far from optimized [70]. Instead of dissipating the regenerative energy as heat, several researchers have proposed to use electrical energy storage elements (ESEs) such as ultra-capacitors and batteries to enable energy recovery [21, 22, 71]. As shown in Figure 2.12 [24], separate electric actuator bus can also be created with ESEs to minimize the impact of the high transient power flow from the actuators and protect the main electrical power network from disturbance and risks [22]. Nonetheless, the ESEs jointly with multiple bidirectional power converters still require considerable additional hardware installment. Using the engine, APU and their electrical starter/generator rotors as fly-wheels to absorb the regenerative energy can take advantage of existing components in the aircraft for mechanical energy storage. This type of solutions demand minimum hardware installment [71]. How- ever, this solution requires the regenerative power flow through the primary distribution,

26 Engine Engine APU

Engine APU Engine S/G 1 S/G S/G 2

Integrated Integrated Integrated control center control center control center (ICC) 1 (ICC) APU (ICC) 2

Separated Other Primary Bus L Primary Bus R Other Separated actuator loads 270 Vdc 270 Vdc loads actuator DC bus DC bus

Figure 2.12. A potential electrical power system architecture with separated actuation DC bus [24] which adds risks to the main DC buses and can potentially increase the copper losses in the wiring. Additional electromagnetic interference (EMI) ﬁlter may also be required.

2.3 Future trends and expectations for the next generation more electric aircraft Architectures

The performance goals declared by NASA in terms of reductions of emissions, noise, and energy consumption is shown in Table 2.2 [7]. Because of these ambitious targets, the interest of electriﬁcation of aircraft system is not conﬁned to non-propulsive power systems. The concept of hybrid electrical propulsion for aircraft system has attracted increasing public attention since late 2000’s [72, 73, 74, 75, 76]. The possible future aircraft hybrid electrical propulsion architectures can be attributed into two categories: the hybrid electric propulsion and the hybrid turbo electric propulsion. A schematic diagram of hybrid electric propulsion is shown in Figure 2.13 [72]. In hybrid

27 Table 2.2. NASA goals for future subsonic aircraft [7]

Generation Noise LTO NOx Cruise NOx Energy (below Stage 4) emission emission consumption N+1 (2015 - 2025) 22 - 32 dB 70 - 75% 65 - 70% 40 - 50% N+2 (2025 2035) 32 - 42 dB 80% 80% 50 - 60% N+3 (beyond 2035) 42 - 52 dB >80% >80% 60 - 80% Note: Goals may vary by aircraft size and mission. Goals for N+1 and N+3 are referenced to Boeing 737-800, whereas goals for N+2 is referenced to Boeing 777.

2 Electric 3 Battery/ Electric Turbo Engine Fuel Cell Network Motor

Fuel

Figure 2.13. Schematic diagram of hybrid electric propulsion architecture electric propulsion architecture, an electric motor is installed on the gas turbine engine shaft to assist the turbine engine or even drive the propulsor device by itself for a small duration of the flight mission [73, 74]. Compared to the conventional Brayton cycle propulsion system, the electrical propulsion system presents a much higher efficiency (91% v.s. 42%). As a result, the addition of electric powered propulsion can benefit the overall propulsion system efficiency (estimated above 50% overall efficiency). However, the mechanical input from the electric motor on the turbine shaft can influence the operation of the gas turbine dramatically.

Rigorous investigation of the interaction between the electric motor and existing gas turbine system is required. In addition, the battery/fuel cell packs add extra weight to the system, architectural considerations and trade-oﬀs of designing the new hybrid electric propulsion power system need to be evaluated in terms of system volume, weight, eﬃciency, reliability, etc [75].

28 3 Electric Rectifier Motor

Electric Network

3 Electric Rectifier Motor 2

3 Electric Rectifier Turbo Engine Generator

Fuel

Figure 2.14. Schematic diagram of hybrid turbo electric propulsion architecture [72]

In Figure 2.14 [72], the schematic diagram of hybrid turbo electric propulsion is demonstrated. In this architecture, a turbo electric distributed propulsion (TeDP) structure is adopted [77, 78]. In stead of installing the electric motor to the gas turbine shaft, in the hybrid turbo electric propulsion system, an electric generator is equipped on the gas turbine engine shaft. Electric power generated from the generator is used to provide electrical powered propulsion, non-propulsive electric power, and to charge the aircraft batteries. In order to provide electrical powered propulsion, electric motors are directly integrated to the propulsor shaft. In this type of hybrid system, the operation of conventional Brayton cycle gas turbine is completely decoupled with the operation of electrical system. With proper architectural design and thoughtful control strategies, both the conventional system and electrical system can be operated at their peak eﬃciency [76, 79].

29 Notwithstanding the promising future opportunities of aircraft hybrid electrical propulsion, the key enabling technologies for electriﬁcation of aircraft propulsion are still far from matured. In the meantime, the development of electriﬁcation for non-propulsive power system on aircraft can help advance the existing aircraft subsystems (including engine system,

ECS, wing anti-icing system, actuation system, etc.) in terms of system weight, volume, complexity, reliability, and maintenance requirement. Such advancements can be obtained by reducing or eliminating the centralized hydraulic systems on board and optimizing the continuously growing electrical power system. The expectations of the next generation MEA architectures may fall into the following directions in the near future:

• The need of increasing electrical power generation capability lead to the trend of shar-

ing load power between multiple spools of the main engine. Paralleling generators in

either CF or VF AC primary power generation system is problematic. However, in the

newly emerged DC primary generation system, the AC power supplied to the frequency

insensitive AC loads is transmitted through a two-stage AC-DC-AC conversion. Such

arrangement adds extra losses and additional hardware to the system. Neither AC

nor DC primary generation system is able to meet all the power requirements with

optimized performance in terms of volume, weight, eﬃciency, reliability and cost. Ma-

jor changes in electrical power generation and distribution architecture are required to

combine the advantages and address the shortcomings of both types of systems.

• Substituting hydraulic actuators for electro-hydrostatic or electro-mechanical actuators

requires cautious investigation on the high peak and regenerative power management

issues. Returning the regenerative energy to the power source(s) requires minimum

hardware installment, however, without a separate electric actuation bus, securing the

operation of the main aircraft electrical power grid within the limits of the speciﬁed

standards is quite a challenging task. New power generation and management system

30 architecture is demanded to take advantage of the existing components in the aircraft for mechanical energy storage.

• The electrical separation can be created by additional power electronic devices with ESEs. As one of the most critical load sets, the power supply of the actuation loads must be secured in all times during the flight mission. In case of equipment failure in the bus separation power electronic devices, the highly dynamic actuation loads cannot be directly connected to the primary DC bus. In this scenario, these safety critical electrical loads are isolated from the most of the available power sources on board. Although aircraft batteries can provide emergency power to the actuators, to fulfill the ETOPS requirement and reduce the reserved aircraft battery capacity, redundant power flow path to the primary or secondary electrical power sources should be available for the separated actuation bus.

Considering the excellent re-conﬁgurability of electrical power system, the fault protection and fault-tolerance of the new electrical power system and aircraft critical load power supply are to be enhanced. New architecture for the separated power supply of ﬂight control actuation system is expected.

31 CHAPTER 3

INDUCTION GENERATOR BASED AC/DC HYBRID POWER

GENERATION SYSTEM FOR MORE ELECTRIC ENGINE 12

In MEA system, both AC and DC electric power with multiple voltage levels are required for various aircraft loads. In this chapter, an induction generator based AC/DC hybrid electric power generation system for MEA is presented. In the proposed system, a high speed induction starter/generator and a low speed induction generator are installed on the high pressure (HP) and low pressure (LP) spools of the engine, respectively. In generating mode of operation, all of the constant voltage variable frequency AC power is generated by the HP generator while the DC power demand is shared by both HP and LP generators.

The operation of proposed system is analyzed under both current and rotor ﬂux oriented reference frames. Control schemes based on instantaneous power control theory and ﬁeld oriented control theory are developed to regulate the AC load voltage and coordinate DC power generation between the two generators. The proposed induction generator based

AC/DC hybrid generation system results in reduced hardware requirement compared to existing AC and DC primary generation systems.

3.1 Introduction: AC and DC primary main engine electric power generation

architectures

In current MEA systems, wound ﬁeld synchronous generator based AC primary generation system [52, 16] are widely used for the main engine electrical power generation, the system

1 c 2014 IEEE. Portions Adapted, with permission, from Y. Jia, U. R. Prasanna and K. Rajashekara, “An open-end winding induction generation system for frequency insensitive AC loads in more electric aircraft,” IECON 2014 - 40th Annual Conference of the IEEE Industrial Electronics Society, Dallas, TX, 2014.

2 c 2015 IEEE. Portions Adapted, with permission, from Y. Jia, K. Rajashekara, “An induction generator based AC/DC hybrid electric power generation system for more electric aircraft”, IEEE Industry Applications Society Annual Meeting, Addison, TX, 2015.

32 configuration of which is shown in Figure 3.1 [52, 16]. In this system, a wound field synchronous starter/generator is connected to the HP spool of the gas turbine engine through a mechanical gearbox. For starting the engine, the synchronous machine is operated as a motor to start the gas turbine using ground power supply. During the flight mission, the same machine acts as a generator and serves as the main electrical power source at a constant

AC voltage (230 V) and variable frequency (360 to 800 Hz) [16]. The field current of the synchronous generator is controlled by a smaller PM machine with a diode bridge rectifier installed on the generator shaft. By adjusting the excitation of the field winding, the AC source voltage can be regulated with variable shaft speed. In this way, the WFSG based

AC primary generation system can feed the frequency insensitive loads directly from the synchronous generator terminals. However, the complex rotor structure makes the torque to inertia ratio of WFSG lower than other type of electric machines [8]. Moreover, the rotating diode bridge structure limits the top speed of the generator shaft. If the synchronous machine is used as a starter/generator, separate ﬁeld and armature voltage controls are required during its motoring operation.

In aircraft systems, the effect of electrical power offtake can sometimes have significant impact on the dynamics and control of the aircraft engine. For instance, during the transition from cruise to descent phase, the aircraft engine power is transiently reduced while maintaining high electrical power demand. This transition creates a possibility of engine instability and may require substantial electric load shedding. Furthermore, with the increasing electric power consumption in MEA, the above effect will be more severe if the electric power is solely extracted from the HP spool of the gas turbine engine [29]. This issue can be resolved by installing an extra generator on the LP spool of the engine and sharing the power extraction between the HP and LP spools [30]. In this way, the power generated from the LP spool could compensate for the decreased power from the HP spool so that the electrical power demand is not compromised.

33 Synchronous Gas Turbine Gearbox Starter/ HP Spool Generator

3-Phase VF (360 – 800 Hz) 230 V AC Distribution

Auto Transformer Transformer Auto Transformer Rectifier Unit Rectifier Unit Unit

± 270 V DC Distribution 28 V DC Bus 115 V AC Distribution

Figure 3.1. System conﬁguration of a wound ﬁeld synchronous generator based AC primary generation system [52, 16]

The idea of adding a fan-shaft generator on the low pressure spool has been widely accepted with the concept of MEE [29, 30, 31]. In a twin-spool aircraft engine, the generators on HP and LP spool operate at different frequencies. In order to parallel the two generators with enhanced efficiency and reduced size and weight, a DC primary generation system with power electronic converters is preferred as an advanced more electric architecture [56, 18, 19, 20]. PM generator has been investigated for this twin-spool twin-generator architecture due to its high power density and self-excited capability [56, 18]. As shown in Figure 3.2, a high speed starter/generator and a low speed generator are placed directly on the HP and LP spool of the engine, respectively. In the engine starting process, the PM starter/generator on HP spool can operate as a motor to start the engine using ground power supply. During flight mission, the power generated from the two generators are rectified and transmitted to a +/-270 V DC power bus. However, the amount of electrical power consumption of

34 HP PM Gas Turbine Gas Turbine Starter/ LP PM HP Spool LP Spool Generator Generator

Active Rectifier Active Rectifier Unit Unit

± 270 V DC Distribution

Auto Transformer or DC/DC PWM Inverter Unit PWM Inverter Unit Converter Unit

230 V AC Distribution 115 V AC Distribution 28 V DC Bus

Figure 3.2. A potential system conﬁguration of a DC primary generation system [56] frequency insensitive AC loads is still signiﬁcant. In Boeing 787, power consumption of frequency insensitive AC loads under cursing condition is close to 50% [57]. This amount of

AC power is supplied through a two-stage AC-DC-AC conversion. Such arrangement adds extra losses and additional hardware to the system.

In AC primary generation system, the frequency insensitive loads are powered directly from the synchronous generator terminals, where the AC load voltage is regulated by adjusting the excitation of the ﬁeld winding of WFSG according to the shaft speed variation. A new approach of generating CVVF AC power directly from the generator terminals without external excitation is required in the new power generation system under MEE concept.

35 Current Filter for Shunt Connected Converter

ea Ls Rs Lf e b Ls Rs Lf C ec Ls Rs Lf

Shunt Connected VAR Compensator Induction Generator CVVF AC Loads

Figure 3.3. Circuit diagram of conventional shunt connected induction generation system [80]

3.2 Direct AC power generation architectures for frequency insensitive loads

on high pressure spool

In order to directly supply CVVF power from the generator terminals, the frequency insensitive loads can be either shunt connected or series connected to the generator terminals. As a fail-safe self-excited electric machine, induction machine is selected to be the new main engine starter/generator to generate CVVF power directly from the generator terminals, and also to supply constant voltage DC power from the same generator.

The circuit diagram of a conventional shunt connected induction generation system is shown in Figure 3.3 [80]. In this configuration, the CVVF loads are directly connected to the generator terminals. A volt-ampere reactive (VAR) compensator is connected in parallel with the CVVF loads through a passive current filter. This current filter at converter output terminals in this configuration requires extra hardware footprint compared to the series connected inverter-load topology [81]. Because the generator terminals and CVVF loads are connected in parallel, the current rating of the generator is the sum of the current ratings of the AC and DC loads.

36 Open-End Winding Induction Generator

e Battery for Input a Ls Rs Power Compensation

eb Ls Rs

ec Ls Rs

CVVF AC Series Connected Loads VAR Compensator

Figure 3.4. Inverter-load topology in the battery compensated series connected induction generation system [81]

The inverter-load topology in [81] for a battery compensated generation system is shown in Figure 3.4. This generation system is used for alternate power generation (e.g., wind energy systems) in isolated areas. The battery at DC side is used to compensate ﬂuctuation in mechanical input power from the alternate energy source. Compared to the conventional shunt connected conﬁguration, the series connected VAR compensator requires higher current ratings. However, since the open-end winding generator can supply the same amount of power required by the AC and DC loads with higher voltage but lower current, the size and weight of the generator can be greatly reduced compared to the shunt connected con-

ﬁguration.

The inverter-load topology is adopted for CVVF power generation on HP spool of the aircraft engine and shown in Figure 3.5. In the proposed aircraft engine generation architecture, an open-end winding induction starter/generator is used to supply both AC and

DC power to the series connected generation system. A three-phase inverter/rectifier unit and the frequency insensitive AC loads are connected to each ends of the open-end winding induction generator terminals, and a DC link capacitor is used on the DC side of the inverter/rectifier unit. The series connected generation configuration for direct CVVF power generation enables a new AC/DC hybrid power generation architecture for the main engine

37 Open-End Winding Induction Generator

ea Ls Rs

eb Ls Rs C CV DC Loads ec Ls Rs

CVVF AC Inverter/Rectifier Unit Loads

Figure 3.5. System conﬁguration of the proposed Open-end Winding Induction Generation System generation system in MEA. The overall AC/DC hybrid power generation on both HP and

LP shaft of the more electric engine is discussed in the following section.

3.3 Proposed AC/DC hybrid power generation architecture

The overall system conﬁguration of the proposed induction generator based AC/DC hybrid generation system is shown in Figure 3.6. Similar to the DC primary generation system, the +/-270 V DC power demand is shared by two generators on HP and LP spool of the engine. In contrast, the frequency insensitive AC loads are supplied directly from the HP spool generator terminal like the AC primary generation system. Compared to DC primary generation system in Figure 3.2, the undesired AC-DC-AC conversion is avoided by applying

AC/DC hybrid generation on HP spool. As compared to the AC primary generation system in Figure 3.1, the application of induction generator removes the external exciter, while the twin-spool twin-generator architecture improves the overall generation performance.

A more detailed electrical system configuration is shown in Figure 3.7. An inverter/rectifier unit and frequency insensitive AC loads are connected to each end of the HP spool open-end winding induction generator terminals. An active rectifier unit is connected to the LP spool

38 HP Induction Gas Turbine LP Induction Gas Turbine Starter/ HP Spool Generator LP Spool Generator

Active Active Rectifier Rectifier Unit Unit

115 V AC Distribution ± 270 V DC Distribution

Auto Transformer DC/DC Unit Converter Unit

230 V AC Distribution 28 V DC Bus

Figure 3.6. System configuration of the induction generator based AC/DC hybrid generation system wye-connected induction generator. The DC output end of the inverter/rectifier unit and the active rectifier unit are paralleled to the DC bus.

As mentioned in earlier sections, in most of the MEA applications, besides generating electric power, the main engine generator is also used as a starter for starting the aircraft engine. A DC power supply from the APU generation system or ground power supply is usually available for this process. As is shown in Figure 3.8, in the engine starting mode of operation, the entire LP generation subsystem is deactivated. The AC loads are disconnected from the HP generator, and the AC load side generator terminals are shorted to transform the open-end induction generator on HP spool into a wye-connected induction motor. Using the

DC power supply, the transformed induction motor can be driven by the inverter/rectiﬁer

39 Open-end Winding Induction External DC Generator On HP Spool Power Inverter/ e Rectifier Unit a Ls Rs e b Ls Rs e c Ls Rs

CVVF AC Loads

Active e Rectifier Unit a Ls Rs

eb L R s s CV DC C e Loads c Ls Rs

Wye-connected Induction Generator On LP Spool

Figure 3.7. Electrical system diagram of the induction generator based AC/DC hybrid generation system unit to start the aircraft engine. Once the engine shaft reaches its idle speed, induction machine will be connected back to form the conﬁguration as shown in Figure 3.9, and the

DC capacitor will be fully charged. Additional circuit breakers are required to implement this transformation. In generator mode of operation, all the CVVF power is generated by the HP generator only, while the power demand of the DC loads is shared between both the HP and

LP generators. As shown in Figure 3.9, in the proposed generation system, the generation subsystem on HP spool includes a high speed generator, an inverter/rectiﬁer unit, a CVVF distribution bus, and a shared CV DC distribution bus, whereas the generation subsystem on LP spool consists of a low speed generator, an active rectiﬁer unit, and the shared CV

DC distribution bus. The system modeling and operating principle of the two generation subsystems are explained in the following sections.

40 Open-end Winding Induction External DC Generator On HP Spool Power Inverter/ e Rectifier Unit a Ls Rs ia e b Ls Rs ib vab e c Ls Rs ic vbc

CVVF AC Current Loads Feedback

Controller 1

Active e Rectifier Unit a Ls Rs

eb L R s s CV DC C e Loads c Ls Rs

Wye-connected Induction Generator On LP Spool

Figure 3.8. Starter mode of operation of the induction generator based AC/DC hybrid generation system

3.4 High pressure spool generation subsystem modeling and operating principle in generating mode: instantaneous power control theory based approach

In the HP spool generation subsystem, the frequency insensitive loads in MEA such as wing de-icing system and galleys, which are generally resistive heaters, are symmetrically distributed at the generator terminals. Therefore, the open-end winding induction generator on HP spool can be modeled as a wye-connected induction generator with an increased stator resistance. In the synchronously rotating reference frame, assuming balanced impedance in both stator and rotor circuit and neglecting the saturation eﬀects, the voltage equation for three-phase squirrel-cage open-end winding induction generator with series connected

41 Open-end Winding Induction HP spool generation subsystem Generator On HP Spool Inverter/ e Rectifier Unit a Ls Rs ia e uab b Ls Rs ib vab e ubc c Ls Rs ic vbc

CVVF AC Current Loads Feedback

AC Voltage Feedback Controller 1 D

DC Power Command C

V o l t a

Controller 2 g e

F e e

Current d b

Feedback a c e k a Ls Rs ia

eb L R ib vab s s CV DC vdc C e Loads c Ls Rs ic vbc

Active Rectifier Unit Wye-connected Induction Generator On LP Spool LP spool generation subsystem

Figure 3.9. Generator mode of operation of the induction generator based AC/DC hybrid generation system resistive load can be expressed as [82]: dλ v = R T i + ω λ + qs (3.1) qs s qs e ds dt

dλ v = R i − ω λ + ds (3.2) ds s ds e qs dt

dλ 0 = R i + (ω − ω ) λ + qr (3.3) r qr e r dr dt

dλ 0 = R i − (ω − ω ) λ + dr (3.4) r dr e r qr dt

42 where

λqs = Llsiqs + Lm (iqs + iqr) (3.5)

λds = Llsids + Lm (ids + idr) (3.6)

λqr = Llriqr + Lm (iqs + iqr) (3.7)

λdr = Llridr + Lm (ids + idr) (3.8)

3 P T = (λ i + λ i ) (3.9) e 2 2 ds qs qs ds In the above equations, the s subscript denotes variables or parameters in the stator

circuit. vqs, vds, iqs, ids, λqs, λds are the q and d axis stator voltages, currents, ﬂux linkages respectively. RsT is the total stator resistance, which include the stator winding resistance and AC load resistance. Lls stands for the stator leakage inductance whereas Lm is the magnetizing inductance of the induction machine. The r subscript denotes variables or parameters in the rotor circuit, and the variables and parameters are deﬁned in a similar way in the stator circuit. As a dual to the voltage-oriented reference frame [83, 84], a reference frame in quadrature with stator current vector [81] is utilized to decouple the active and reactive voltage components from the output voltages of series connected converter. Both AC and DC side power outputs of the generation system can be regulated independently by controlling the instantaneous active and reactive power outputs of the series connected converter. In the current oriented reference frame, the direct and quadrature components of the stator currents in the series connected induction generation system are:

43 iqs = |i| (3.10)

ids = 0 (3.11)

In steady state, the derivative terms in Equation 3.1 to Equation 3.4 are equal to zero. Hence, the voltage equation can be re-written as:

Vqs = RsT Iqs + ωeLmIdr = VP s (3.12)

Vds = −ωe (Ls|I| + LmIqr) = VQs (3.13)

0 = RrIqr + ωslLr|I| (3.14)

0 = RrIdr − ωsl (LrIqr + Lm|I|) (3.15)

where

Ls = Lls + Lm (3.16)

Lr = Llr + Lm (3.17)

ωsl = ωe − ωr (3.18)

Equations 3.12, 3.13, 3.14, and 3.15 describe the steady-state behavior of the open-end winding induction generator for series connected CVVF AC generation system, where VP s is the active voltage component which is in phase with the stator current vector, and VQs is the

44 reactive voltage component that leads the stator current vector by 90 degree. Substituting Equations 3.14 and 3.15 into Equations 3.12 and 3.13, the active and reactive voltage components of the induction generator terminal voltages can be expressed as:

ω R L 2 V = R + ω sl r m |I| (3.19) P s sT e 2 2 2 Rr + ωsl Lr

L ω 2L 2 V = −ω L − r sl m |I| (3.20) Qs e s 2 2 2 Rr + ωsl Lr

By changing the reactive voltage output of the series connected converter VQs, the excitation of the HP generator is controlled. The generator stator current magnitude can then be regulated according to the AC load variation. By changing the active voltage output of the converter VP s, the power delivered to the DC bus can be controlled.

3.5 High pressure spool generation subsystem modeling and operating principle in generating mode: ﬁeld orientation control theory based approach

The analysis based on instantaneous power theory [85] shows that the HP spool generation subsystem can be regulated by controlling the power flow between the generator and inverter/rectifier unit. However, the flux and electromagnetic torque of HP generator cannot be directly regulated under current orientation. In order to decouple the AC and DC side power generation by directly controlling the HP generator, a field oriented control theory based analysis of the HP spool generation subsystem is investigated in this section. In the rotor flux oriented reference frame, neglecting the saturation effects, the steady- state voltage and torque equations for the HP spool induction generator can be expressed as [82, 86]:

Vqs1 = (Rs1 + RacL) Iqs1 + ωe1λds1 (3.21)

Vds1 = (Rs1 + RacL) Ids1 − ωe1λqs1 (3.22)

45 2 3 P 1 Lm1 Te1 = Iqs1Ids1 (3.23) 2 2 Lr1 Where,

λqs1 = Lsσ1Iqs1 (3.24)

λds1 = Ls1Ids1 (3.25)

2 Lm1 Lsσ1 = Ls1 − (3.26) Lr1

In the above equations, Vqs1, Vds1, Iqs1, Ids1, λqs1, λds1 are the q and d axis stator voltages, currents, ﬂux linkages, respectively. Rs1 is the stator winding resistance and RacL is the

AC load resistance. Lsσ1 stands for the stator transient inductance (stator short-circuit inductance). Ls1, Lr1, Lm1, P1, are the stator, rotor magnetizing inductance and pole pairs of the HP generarator (induction machine 1), respectively. In order to regulate the AC load voltage, the AC load current (stator current magnitude) needs to be controlled according to the load resistance variation. The AC load current reference can be expressed as:

q ∗ 2 2 I = Iqs1 + Ids1 (3.27)

Limited by the current rating of the generation system, the power generated for frequency insensitive AC loads can be modeled as the ohmic loss of the increased stator resistance of the HP generator. Thus, the power transmitted to the DC bus can be written as:

T ∗ P = −ω e1 − (R + R ) I∗2 (3.28) dc1 e1 P 1 s1 acL

∗ where Te1 is the torque reference of the HP generator.

46 300

200 Voltage Constraint

100 AC Load Current Reference

0 Two Theoretical

iqs (A) Equilibrium Points

Operating Range of DC -100 Output Power for Given AC Load Condition B

-200

A Torque Reference -300 -300 -200 -100 0 100 200 300 ids (A)

Figure 3.10. Operating constraints of the high pressure spool generation subsystem

According to Equation 3.27, the torque reference can be determined by the AC load current reference and DC power output command. The theoretical equilibrium points of the

HP generation subsystem for a given AC and DC power demand are illustrated in Figure 3.10.

In Figure 3.10, the intersections A and B indicate two theoretical equilibrium points for corresponding AC load current and HP generator torque references. For a given AC load condition, the torque reference of the HP generator varies with diﬀerent DC output power command of the system. When the DC power output increases, the torque reference curve will move away from the AC load current reference circle. Therefore, the DC power output

47 needs to be limited to ensure that there exists at least one intersection point of the torque reference curve and the load current reference circle. Furthermore, the maximum torque of an induction generator is limited by the voltage rating of the system. This voltage limitation can be expressed as follows:

2 2 2 Vs1,max ≥ Vqs1 + Vds1 (3.29)

Substituting Equations 3.21, 3.22, 3.24, 3.25 into Equation 3.29, the voltage constraint equation becomes:

2 (Rs1 + RacL) Iqs1 2 (Rs1 + RacL) Ids1 2 Vs1,max + LsIds] + − Lsσ1Iqs1] ≤ 2 (3.30) ωe1 ωe1 ωe1

Equation 3.30 forms a voltage limit ellipse in Figure 3.10. This ellipse is similar to the analysis of ﬂux weakening operation for induction motor [86], yet the load resistance makes the ellipse rotation varies from diﬀerent load condition in the proposed system. An increased AC load power demand will result in a clock-wise rotation of the voltage limit ellipse and vice versa. Therefore, equilibrium point B in Figure 3.10 is not a valid operating point of the system. Besides the constraint to guarantee the existence of equilibrium point, the DC output power from HP spool is hereby further bounded by the voltage limit of the system. This bounded operating range can be demonstrated as the segment of the AC current reference circle inside of the voltage limit ellipse.

The ﬂux current command Ids,ref and torque current command Iqs,ref can be calculated from given AC load current and generator torque reference as follows:

q ∗ q ∗ ∗2 2Te1 ∗2 2Te1 I − k − I + k I∗ = 1 1 (3.31) ds1 2

q ∗ q ∗ ∗2 2Te1 ∗2 2Te1 − I − k − I + k I∗ = 1 1 (3.32) qs1 2

48 Iqs RIqse reff contourerence Ids RIde ref fcontourerence ) ) . . u u . . 0.7 0.7

p 0.7 p 0.7 ( (

e e c c

n 0.60.6 n 0.60.6 e e r r e e f f e e 0.5 0.5 R 0.5 R 0.5

e e u u q q

r 0.40.4 r 0.40.4 o o T T

c c i i

t 0.3 t

0 (p.u.) Electromagnetic Torque .3 0.30.3

e e Electromagnetic Torque (p.u.) Electromagnetic Torque n n g g a a 0.20.2 0.20.2 m m o o r r t t c c 0.1 e 0.1 e 0.10.1 l l E E 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 Current Magnitude (p.u.) 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.1 0.2 0.3 0Current.4 Magnitude0.5 (p.u.)0.6 0.7 0.8 AC Load Current Reference (p.u.) AC Load Current Reference (p.u.)

Figure 3.11. Contour maps of d-q axis current commands for AC load current and generator torque reference sweep test

2 3 P 1 Lm1 where k1 = is the torque coefficient. 2 2 Lr1 In order to illustrate the disposition of d, q axis current commands with AC and DC power demand variations, a two-dimensional sweep test of AC load current and generator torque reference is performed using MATLAB. The results of the test are shown as contour maps in Figure 3.11. Evidently, the q-axis (torque) current command and AC load current reference variations are almost proportional, while changing q-axis current command has little impact to the generator torque with a fixed AC load current command. The d-axis (flux) current command changes greatly with generator torque variation, but it is relatively insensitive to AC load current changes. It can be inferred from the sweep test results that changing DC power output of the HP generation subsystem greatly depends on the flux variation of HP generator, resulting in a very slow system response. In contrast, regulating AC load voltage with a constant (or slowly varied) DC output power can be achieved by controlling torque current with small variation of generator flux. In rotor flux oriented reference frame, the DC bus voltage may be regulated with a master-slave control strategy with cooperation of the LP generation subsystem.

49 3.6 Low pressure spool generation subsystem operating principles and DC volt-

age regulation

The generation subsystem on LP spool consists of a low speed generator, an active rectiﬁer unit, and the shared CV DC distribution bus. The LP spool generation subsystem controller is responsible for the DC voltage regulation of the proposed hybrid AC/DC generation system.

The DC voltage regulation of the whole generation system is implemented by a master- slave control strategy. The HP generator, while regulating AC load voltage, operates as the master in DC power generation and oﬀers a constant power output to the DC bus. In contrast, the LP generator operates as the slave and supplies the rest of DC power demand.

The DC power output of the LP generator is determined as:

dv P = v C dc + v i − P (3.33) dc2 dc dt dc dcL dc1

where idcL is the DC bus load current, Pdc1 and Pdc2 are the DC power output of the HP and LP generator, respectively.

Besides the DC power extraction from the HP generation subsystem, the LP spool generation subsystem operates as a conventional ﬁeld oriented controlled front end DC generation system.

3.7 Closed-loop control scheme for generating mode of operation

The overall closed-loop control scheme for the proposed AC/DC hybrid generation system is shown in Figure 3.12. The AC load voltage is solely controlled by the HP generator controller. The DC bus voltage is regulated by a master-slave control strategy which involves both the HP and LP generator controllers. The DC power output command from HP generator controller (master controller) is designed as a control input commanded by the

50 CVVF AC pac pdc1 (Master) To Main pdc2 (Slave)

Load DC Bus

. . HP LP IG IG

* v vdc * ac ia1 ib1 V1 V2 ia2 ib2 * Vac * * HP Gen. LP Gen. Vdc Pdc1 Controller Controller

vac ia1,b1 vdc ia2,b2

Figure 3.12. Overall closed-loop control scheme for the proposed AC/DC hybrid generation system engine control unit in order to prevent instability issue of the engine caused by high oﬀ-take of power extraction on HP spool. An additional control freedom is provided to the engine control unit to balance the power extraction from the HP and LP spools. The LP generator controller (slave controller) is commanded to fulﬁll the rest of the DC power demand from the main DC bus.

Two voltage sensors are used at the AC load terminals of the generator to monitor the

AC load voltages. An additional voltage sensor is installed to measure the DC bus voltage.

Four current sensors are used to provide generator current feedbacks. Neither rotational speed of rotor nor rotor position feedback is essential to control the proposed generation system. The control scheme of both the HP and LP generation subsystems are discussed in detail in the following subsections.

51 3.7.1 Instantaneous power control theory based control scheme for high pres-

sure spool generation subsystem

The closed-loop control scheme developed based on instantaneous power control theory [85] for the HP spool generation subsystem is shown in Figure 3.13.

In order to independently control the AC load voltage and DC power output of the HP generation system, the proposed control scheme is operated under stator current oriented synchronous reference frame. In the current oriented control for induction generator, the d axis voltage (reactive voltage component) regulates the instantaneous reactive power output

To main DC bus

* * Vac PI vds * vabcs Controller - Inverter/ dq SVM abc rectifier Unit * * Pdc1 vqs Divider θi

i , i PLL a b

i q dq HP abc IG

vac

CVVF AC Load

Figure 3.13. Closed-loop current oriented control scheme for the HP spool generation subsystem

52 of the converter which is controlling the excitation for the induction generator, while the q axis voltage (active voltage component) controls the instantaneous active power output of the converter that provides real power to the DC bus.

∗ Assuming resistive load condition, the AC load voltage magnitude Vac is proportional to the stator current magnitude |I|. According to Equations 3.10, 3.11 and 3.19, for a given

steady-state operating point, the current magnitude (|I| = Iqs) can be determined solely by the d axis voltage (reactive voltage) command. Therefore, a PI controller is used to regulate

∗ ∗ Vac by controlling the reactive voltage component of the converter output VQs.

∗ The DC power output command Pdc1 is designed as a control input given by the engine

∗ control unit. The q axis voltage (active voltage) command VP s can be obtained as:

2P ∗ V ∗ = dc1 (3.34) P s 3|I| In the current oriented control, a conventional phase locked loop (PLL) is used to obtain

the stator current vector phase angle θi. The basic conﬁguration of the conventional PLL is shown in Figure 3.14 [87]. In order to make the current oriented reference frame in quadrature with stator current vector, the direct current component id is commanded to be

zero. The current vector phase angle θi is obtained by integrating the frequency command

∗ ωe generated by the id controller. As shown in Figure 3.13, the output of this PLL is used

* * ω θi id =0 + PI e Integrator Controller -

id dq i q abc ia, ib, ic

Figure 3.14. Basic conﬁguration of conventional phase locked loop

53 in the inverse Park’s transformation to provide the three phase reference voltage signals to space vector PWM modulator.

3.7.2 Field oriented control theory based control scheme for high pressure spool generation subsystem

The control scheme based on instantaneous power theory [85] regulates the HP spool generation subsystem by controlling the power flow between the generator and inverter/rectifier unit. However, the flux and electromagnetic torque of HP generator are not directly controlled. In this section, a field oriented control theory based closed-loop control scheme of the HP spool generation subsystem is developed to decouple the AC and DC power generation by directly controlling the HP generator. The proposed field orientation based control scheme for HP spool generation subsystem is shown in Figure 3.15. In the HP spool generation subsystem, AC and DC power are generated from two sides of the HP generator winding, respectively. The operating range is calculated using Equations 3.27, 3.28 and 3.30, and fed back to the engine control unit.

LP Generation Subsystem

HP Gen. Controller ...To Main DC Bus * * Pdc1 Iqs1 + Current Controller Torque * Current - dq Te1 SVM Reference Reference abc * * Calc. Calc. * Vac I Ids1 + Current θrf1 ÷ - Controller dq ia1, ib1 HP ωe1 abc IG v RacL AC Load ac Estimation CVVF AC HP Generation Subsystem Load

Figure 3.15. Closed-loop rotor ﬂux oriented control scheme for the HP spool generation subsystem

54 In order to decouple the AC and DC power supply of the HP generation subsystem, the

∗ two subsystem control inputs, AC load voltage reference Vac and DC output power reference

∗ Pdc1 , are converted into d and q axis current commands of the HP generator. This conversion can be implemented in four steps:

∗ 1. Given the reference AC load voltageVac , a PI controller is used to generate the AC load current command I∗.

2. The AC load resistance RacL is estimated as the AC load voltage vac divided by the stator current magnitude I.

∗ ∗ 3. Given DC power output command Pdc1 , AC load current command I , AC load re-

sistance RacL and generator electrical frequency ωe1, the torque reference of the HP

∗ generator Te1 is calculated using Equation 3.28;

∗ ∗ 4. Given the torque reference of the HP generator Te1 and AC load current command I ,

∗ ∗ the d and q-axis current commands Ids1 and Iqs1 are obtained from Equations 3.31 and 3.32.

The current control loop of the proposed control scheme is based on ﬁeld oriented control

(FOC) theory. A direct flux observer from [88] is used to provide the rotor flux speed and angle information. This flux observer requires terminal voltage and generator stator current feedbacks. Since the inverter terminal voltages can be re-constructed using DC bus voltage feedback and inverter gating signals, only stator current sensor and DC bus voltage sensor are needed for field orientation. Using rotor-flux orientation, two PI controllers are used in the current control loop to regulate d-axis current ids1 and q-axis current iqs1of the HP generator. Space vector pulse width modulation (SVPWM) is used to generate gating signals for the active rectifier unit.

55 3.7.3 Control scheme for low pressure shaft generation subsystem

The closed-loop control scheme for LP shaft generation subsystem is shown in Figure 3.16.

In the DC bus voltage control loop, a PI controller is used to generate LP generator q-axis

* (torque) current command Iqs2. Since LP shaft generator has a wide speed operating range,

ﬂux weakening operation is required in the LP generation subsystem. The d-axis (ﬂux) current of the LP generator is commanded to be inversely proportional to the generator shaft speed to obtain a wide speed operation range. A speed estimator [88] is used in the

LP generation subsystem to provide shaft speed feedback to the ﬂux weakening operator.

Similar to the current control loop in HP generation subsystem, two PI controllers are used to regulate d-axis current ids2 and q-axis current iqs2 of the LP generator. Space vector pulse width modulation (SVPWM) is used to generate gating signals for the active rectiﬁer unit.

LP Generation Subsystem LP Gen. Controller dq ia2, ib2 LP abc IG * * - Vdc + Voltage Iqs2+ Current θrf2 - Controller Controller dq SVM abc Flux * - I Current ωr2 Weakening ds2+ Controller ...To Main Operator DC Bus vdc

HP Generation Subsystem

Figure 3.16. Closed-loop rotor ﬂux oriented control scheme for the LP spool generation subsystem

56 3.8 Computer simulation results for instantaneous power control theory based control scheme

A case study for the proposed instantaneous power control theory based control scheme is simulated in MATLAB/Simulink. In this study, a 37 kW, 1800 rpm open-end winding induction generator is controlled to supply a 30 kW three phase 230 VAC balanced resistive load. The mechanical rotational speed of the generator shaft is 2000 rpm. No electrical power is transmitted to the DC bus. A DC link capacitor is connected at the DC side of the converter. The preliminary simulation results are shown in Figure 3.17, 3.18, 3.19, 3.20, 3.21, and 3.22. In the simulation, the CVVF load is changed from 30 kW to 15 kW at 0.4 s. Variation in DC and AC side voltages of the system are shown in Figure 3.17 and Figure 3.18 respectively, and characteristics of generator stator currents and electro-magnetic torque are illustrated separately in Figure 3.19 and Figure 3.20. The dynamic performance for both DC capacitor voltage and AC load voltage regulation is found to be satisfactory. However, the interaction

510 ) V (

e 505 g a t l o v

s u

b 500

C D

495 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7 Time (s)

Figure 3.17. The DC link capacitor voltage characteristics of the proposed HP spool generation system controlled by current oriented control scheme

57 290 ) s 270 e m g r

a , t l V 250 ( o

v e

d d u a

t 230 o i l

n C g a A 210 m

190 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7 Time (s)

Figure 3.18. The AC load voltage characteristics of the proposed HP spool generation subsystem controlled by current oriented control scheme

80 Phase A Current (ia) ) Phase B Current (ib)

A 60

( Phase C Current (ic)

n 40 e r r 20 u C

e 0 s a -20 h P

e -40 e r

h -60 T -80 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7 Time (s)

Figure 3.19. The three phase current characteristics of the proposed HP spool generation subsystem controlled by current oriented control scheme

between the reactive power output of the converter and the AC load voltage is clearly nonlinear.

58 ) -30 m N (

e -50 u q r o

T -70

c i t e -90 n g a m

- -110 o r t c e

l -130 E 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7 Time (s)

Figure 3.20. The electromagnetic torque characteristics of the HP spool generator in proposed HP spool generation subsystem controlled by current oriented control scheme

) 0 V (

-50 e

g -100 a t l -150 Q-axis Voltage (vQ) o P-axis Voltage (vP) V

s -200 x i

a -250

, -300 P -350 -400 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7 Time (s)

Figure 3.21. The P and Q axis voltage characteristics of the HP spool generator in proposed HP spool generation subsystem controlled by current oriented control scheme

The d and q axis currents and voltages are shown in Figure 3.21 and Figure 3.22. The steady-state d axis current is maintained as zero indicating the eﬀectiveness of the PLL,

59 70 Q-axis Current (iQ)

P-axis Current (iP)

) 60 A (

t 50 n e

r 40 r u

C 30

s x

i 20 a

Q 10 , P 0 -10 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7 Time (s)

Figure 3.22. The P and Q axis current characteristics of the HP spool generator in proposed HP spool generation subsystem controlled by current oriented control scheme whereas the steady-state q axis voltage is kept to be roughly zero demonstrating that no instantaneous active power is transferred to the DC side of the series connected converter.

3.9 Computer simulation results for ﬁeld orientation control theory based con-

trol scheme

A computer simulation case study for the proposed induction generator based AC/DC hybrid generation system using FOC based control scheme is conducted in MATLAB/Simulink. In this case study, a 150 kW, 12000 rpm induction generator and a 60 kW, 3150 rpm induction generator are used on the HP and LP spool, respectively. The generators are controlled to supply 75 kW three phase 230 V AC balanced resistive load and 60 kW 540 V DC load at their rated speed. The AC load is changed from 75 kW to 60 kW at 1.3s, and the DC load is changed from 60 kW to 50 kW at 1.45s.

60 552552

550550

548548 ) V ( 546546 e g a t l 544544 o v

s 542 u 542 b

D 540540

538538

536536

534534 0.71 10.8.1 10.9.2 1.13 11.1.4 11.2.5 11.3.6 11.4.7 11.5.8 Time (s)

Figure 3.23. The DC bus voltage regulation characteristics of the proposed AC/DC hybrid generation system

145145

140140 ) S

M 135135 R

, V (

e 130

g 130 a t l o v 125125 d a o l

C 120120 A

115115

110110 0.71 10.8.1 10.9.2 1.13 11.1.4 11.2.5 11.3.6 11.4.7 11.5.8 Time (s)

Figure 3.24. The AC load voltage regulation characteristics of the proposed AC/DC hybrid generation system

61 -60-60

-70-70 )

m -80-80 N (

e -90

u -90 q

r Electromagnetic torque

o -100

t -100 of HP generator c i t -110 e -110 n g

a -12-1200 m o

r -130 t -130 c e l

E -14-1400 Electromagnetic torque -15-1500 of LP generator -16-1600 0.71 10.8.1 10.9.2 1.13 11.1.4 11.2.5 11.3.6 11.4.7 11.5.8 Time (s)

Figure 3.25. The electromagnetic torque characteristics of HP and LP generators in the proposed AC/DC hybrid generation system

0.50.5 r o t a

r 0

e 0

n d-axis current e g

P H -0.5

e -0.5 h t

f o q-axis current t

n -1

e -1.0 r r u c

s i x

a -1.-1.55 - q

, d

-2.0-2 0.71 10.8.1 10.9.2 1.13 11.1.4 11.2.5 11.3.6 11.4.7 11.5.8 Time (s)

Figure 3.26. The d, q-axis current characteristics of the HP generator in the proposed AC/DC hybrid generation system

62 0.80.8

0.60.6

0.40.4 ) .

u 0.2

. 0.2

p d-axis current (

t 0 0 n e r

r -0.2

u -0.2 c

s q-axis current i -0.4

x -0.4 a - q

-0.6

, -0.6 d -0.8-0.8

-1.0-1

-1.2-1.2 0.71 10.8.1 10.9.2 1.13 11.1.4 11.2.5 11.3.6 11.4.7 11.5.8 Time (s)

Figure 3.27. The d, q-axis current characteristics of the LP generator in the proposed AC/DC hybrid generation system

The DC bus and AC load voltages of the system are shown in Figure 3.23 and Fig- ure 3.24, respectively. The dynamic performance for both DC bus voltage and AC load voltage regulation is satisfactory.

The electromagnetic torque characteristics of the HP and LP generators are illustrated in Figure 3.25. Clearly, the slow torque response of HP generator is compensated by the fast responded LP generator. At 1.45s, the HP generator operates as the master and does not react as the DC load changes, while the LP generator decreases its torque to accommodate the DC load change.

The d and q axis currents of the HP and LP generators in the proposed system are shown in Figure 3.26 and Figure 3.27. As it is mentioned in Section 3.5, regulating AC load voltage with a constant (or slowly varied) DC output power can be achieved by controlling q-axis current with small variation of generator ﬂux.

63 3.10 Hardware-in-the-loop real-time emulation results

In order to demonstrate the feasibility of operation of the proposed induction generator based AC/DC hybrid generation system, a real time emulation platform is built using OPAL-RT hardware-in-the-loop (HIL) testing system. The emulated generation system is controlled by a Texas Instrument TMS320F28335 digital signal processor (DSP). As is shown in Figure 3.28, the proposed AC/DC hybrid generation architecture is emulated using the OPAL-RT HIL testing system. The proposed architecture includes the HP generator, the inverter/rectiﬁer unit and 115 V balanced resistive AC load; as well as the LP generator, the active rectiﬁer unit and 540 V DC loads on the main DC bus. The OPAL-RT HIL testing platform is consisted of a Virtex 7 FPGA processor and I/O expansion unit (OP5607), a real-time target computer running Redhat operating system (OS), and a PC running the real-time simulation software (RT-LAB). Prior to the real-time emulation, the proposed system model is built and encrypted into FPGA codes in the PC running RT-LAB. The encrypted codes can then be loaded through the target computer running Redhat OS and executed in real-time using the OP5607. The proposed control scheme is implemented in the DSP. The carrier frequencies of the HP and LP generator controllers are set to be 20 and 10 kHz, respectively. The FPGA unit can read the 10 and 20 kHz gating signal through the time-stamped digital input channels of OP5607 and execute the real time emulation with time steps on nanosecond level. However, the feedback analog signals sent from the OP5607 can only update in every 40 µs due to the limited computation power of the target computer. The analog outputs are read by the DSP analog-to-digital converter (ADC) in every switching period. These feedbacks can also be monitored using an oscilloscope through the same analog output channels. A real time emulation case study is conducted under the HIL platform shown in Fig- ure 3.28. In this case study, a 85 kW, 11060 rpm induction generator and a 60 kW, 3150 rpm induction generator are used on the HP and LP spool, respectively. In the beginning of

64 AC/DC Hybrid Generation Architecture OPAL-RT HIL Testing System Inverter/ Active HP Spool To Main LP Spool CVVF AC Rectifier Rectifier Generator DC Bus Generator Load .

Unit . Unit . Ethernet HP LP PC Running RT-LAB IG IG PCIE x4 vdc vac ia1 ib1 ia2 ib2

* * OP5607 Real-time computer V1 V2 * Vac HP Gen. * Analog * LP Gen. Vdc Digital/Analog Output Pdc1 Controller Controller Input/Output

TMS320F28335 Digital Signal vac ia1,b1 vdc ia2,b2 Processor AC/DC Hybrid Generation Controller Monitoring Oscilloscope

Figure 3.28. Hardware-in-the-loop emulation implementation for the proposed AC/DC hybrid generation system the emulation, the generators are controlled to supply 60 kW three phase 115 V AC balanced resistive load and 60 kW 540 V DC load at their rated speed. Step changes are applied to the AC load from 60 kW to 65 kW at 11.0s, and to the DC load from 60 kW to 50 kW at 11.15s. From 10.9s to 11.2s, the HP and LP generator speed ramp to 110% and 120% of their rated speed, respectively. The DC power output command of the HP generation subsystem changes from 20 kW to 10 kW at 11.3s.

Currently, airborne electrical power system does not have an oﬃcial voltage regulation standard for 540 V DC bus. The closest available standard (MIL-STD-704F [68]) is for 270

V DC bus and limits the voltage variation so as not to exceed +10/-20 V in steady-state.

Assuming the voltage variation limitation in MIL-STD-704F is double for the 540V DC bus, the voltage variation allowed for the DC voltage regulation for the proposed system would be +20/-40 V. The transient voltage limitation of the proposed system is assumed to be also double for the same reason. The DC bus voltage waveform is shown in Figure 3.29, the assumed voltage limit of the 540 V DC bus is illustrated in red lines.

The AC load voltage waveform of the proposed system is shown in Figure 3.30, the steady-state and transient voltage limit for a variable frequency 115 V AC distribution in

65 To 660 . . . 575

565 D C

560 b

555 u s

v o

545 l t a g e

535 ( V 525 ) . .

. 515 To 500 10.8 10.9 11.0 11.1 11.2 11.3 11.4 11.5 11.6 11.7 11.8 Time (s)

Figure 3.29. The DC bus voltage regulation characteristics of the proposed AC/DC hybrid generation system

118

108 To 180 . . . 140 A C

130 l o a d

118 120 v o l t 110 a 108 g e

(

100 V r m s

90 )

80 80 A

195 C

l o a

130 d

150 p h

65 a 75 s e

0 0 A

-75 B

-65 v

-150 o l t

-130 a g e

(

-195 V )

10.8 10.9 11.0 11.1 11.2 11.3 11.4 11.5 11.6 11.7 11.8 Time (s)

Figure 3.30. The AC load voltage regulation characteristics of the proposed AC/DC hybrid generation system

66 H P

g e

300 n e r

300 a

200 t o

200 r

100 100 h a s 0 e 0 A

-100 &

-100 B -200

c u

-300 -200 r r e n

-300 t

( A ) 10.8 10.9 11.0 11.1 11.2 11.3 11.4 11.5 11.6 11.7 11.8 Time (s)

Figure 3.31. The phase A and B current characteristics of the HP generator in the proposed AC/DC hybrid generation system

200

100

0 L P

-100 g e

300 n

-200 e r a

200 t o r

100 h a s e

0 A

-100 B

c -200 u r r e n

-300 t

( A ) 10.8 10.9 11.0 11.1 11.2 11.3 11.4 11.5 11.6 11.7 11.8 Time (s)

Figure 3.32. The phase A and B current characteristics of the LP generator in the proposed AC/DC hybrid generation system

MIL-STD-704F is shown as red lines. The third phase AC voltage is absent because only two voltage sensors are installed to monitor the AC load voltage.

The current waveforms of the HP and LP generators in the proposed system are shown in Figure 3.31 and Figure 3.32. The third phase current is absent because only two current sensors are installed for each generator.

67 H P

g e

-40 n e r a t

-50 o r

e l

-60 e c t r -70 o m a g

-80 n e t i -90 c

t o r

-100 q u e

( 10.8 10.9 11.0 11.1 11.2 11.3 11.4 11.5 11.6 11.7 11.8 N m

Time (s) )

Figure 3.33. The electromagnetic torque characteristics of the HP generator in the proposed AC/DC hybrid generation system L P

g e n

-5 e r a t -30 o r

e l -55 e c t r o

-80 m a g

-105 n e t i c

-130 t o r q

-155 u e

( N

10.8 10.9 11.0 11.1 11.2 11.3 11.4 11.5 11.6 11.7 11.8 m Time (s) )

Figure 3.34. The electromagnetic torque characteristics of the LP generator in the proposed AC/DC hybrid generation system

The electromagnetic torque characteristics of the HP and LP generators are illustrated in Figure 3.33 and Figure 3.34. When the DC load decreased from 60 kW to 50 kW at

11.15s, the HP generator operates as the master and does not react as the DC load changes, while the LP generator decreases its torque to accommodate this load variation. At 11.3s, the DC power output command changed from 20 kW to 10 kW. Since the DC power output of the HP generation subsystem greatly depends on the ﬂux variation of HP generator, the torque response of HP generator is much slower compared to its immediate reaction to the

68 AC load change at 11.0s. Clearly, the slow torque response of HP generator is compensated by the fast response of the LP generator. Both HP and LP generators respond to the speed variation from 10.9s to 11.2s. The DC and AC voltages are not aﬀected by the speed variation. The torque characteristics are monitored only for demonstration purposes and no torque transducer is needed for the proposed control scheme.

3.11 Summary

In this chapter, an induction generator based AC/DC hybrid generation system for MEA is presented. The application of induction generator removes the brushless exciter and improves the system power density compared to synchronous generator based generation system. The problem of excessive fault current due to the PM excitation in PM generator based generation system is also avoided. The proposed AC/DC hybrid generation architecture supplies CVVF power directly from generator terminals. As a result, the hardware requirement is reduced compared to both AC and DC primary generation systems. The proposed AC/DC hybrid generation system is analyzed under both current and rotor ﬂux oriented reference frames. Control schemes based on instantaneous power control theory and ﬁeld oriented control theory are developed. Both AC and DC output voltages of the system can be well-regulated with generator speed, AC and DC side load and DC power output command variation using FOC based control scheme. The feasibility of operation of the proposed system is demonstrated by computer simulation and HIL real-time emulation.

69 CHAPTER 4

AN INDUCTION GENERATOR BASED AUXILIARY POWER UNIT FOR

POWER GENERATION AND MANAGEMENT SYSTEM FOR MORE

ELECTRIC AIRCRAFT 1

In this chapter, an auxiliary power unit for power generation and management system to supply/absorb the highly dynamic power demand/regeneration from the EHAd and EMAs is proposed. The proposed system utilizes an open-end winding induction starter/generator

(OEWIS/G) to create a separate DC bus for the actuators without adding signiﬁcant hardware to the system. During the entire ﬂight mission, the regenerative power is recovered by one side of the OEWIG terminals; meanwhile, the power delivery to the main DC network of the aircraft electrical power system can be independently controlled by using the same generator through the other side of the terminals. A closed-loop control scheme based on

ﬁeld orientation control and instantaneous power theory is developed to regulate both the main DC bus voltage and the electric actuation DC bus voltage simultaneously in aircraft emergency power mode.

4.1 Introduction

In the trend towards MEA, the traditional hydraulic ﬂight control actuators with central hydraulic system are being replaced by EHAs or EMAs. Compared to the traditional actuators, EHAs and EMAs have lower weight and volume, increased reliability and reduced maintenance [59, 60, 61]. However, these actuators have highly dynamic power proﬁle, which includes high peak power demand and power regeneration [65, 66, 67].

1 c 2016 IEEE. Portions Adapted, with permission, from Y. Jia, K. Rajashekara, “An induction generator based auxiliary power unit for power generation and management system for more electric aircraft”, IEEE Energy Conversion Congress & EXPO, Milwaukee, WI, 2016.

70 Inverter/ converter unit

ES/G Auxiliary power unit (APU)

Power Loads ES/G Distribution Network Electric drives EHA/EMA Engine Inverter/ converter unit

Figure 4.1. The need of direct power path between the aircraft electrical power source and the highly dynamic actuator loads

The most widely used regenerative power management method for EHA/EMAs on aircraft is to employ resistor banks paralleled with the actuators, dissipating the regenerated energy with associated cooling devices. This solution renders the weight, volume, and ef-

ﬁciency of the overall system far from optimized [12]. Potential regenerative power management methods include using ESEs such as ultra-capacitors and batteries to restore and recover the regenerative energy, and returning the regenerative energy to the power source such as main engine generator or APU generator [21, 22, 71]. Using the energy restoring method with electrical storage elements (ESEs), separate electric actuator bus can be created to protect the main electrical power network from disturbance and risks; however, the

ESEs jointly with multiple bidirectional power converters requires considerable additional hardware installment [22]. Returning the regenerative energy back to the power source has minimum hardware installment requirements, but the returning regenerative power needs to pass through the main electric power network, which could expose the entire aircraft electrical power grid upon risks, such as instability of the bus voltage regulation [71]. As shown in Figure 4.1, in order to secure the operation of the aircraft main electric network, a direct

71 Inverter/ converter unit

Electric drives EHA/EMA Inverter/ converter unit

ES/G Auxiliary power unit (APU)

Power ES/G Distribution Loads Network Engine Inverter/ converter unit

Figure 4.2. The proposed power generation and management system architecture with regenerative power absorption capability power path between the aircraft electrical power source and the highly dynamic actuator loads is desired. The sources that can absorb the regenerative power on the aircraft include the turbine shafts of the main engines and APU engine. However, returning the regenerative power to the main engine shaft is presently prohibited by FAA due to engine operation reliability concerns. As a result, the APU engine turbine shaft becomes a favorable choice as the power source for regenerative power absorption. A power generation and management system architecture that allows the regenerative power from the actuators to be absorbed by the turbine shaft of the APU is proposed and shown in Figure 4.2. In the proposed architecture, the open-end winding topology [89] is adopted to provide direct power ﬂow path from the EHA/EMAs to the power source, and to create a separate electric actuation bus without

72 signiﬁcant additional hardware requirement. Since induction generators present higher power density compared to wound-ﬁeld synchronous generators and better robustness and fault- tolerant capability compared to PM generators in MEA systems [90, 91], an open-end winding induction starter/generator is used as the APU starter/generator. The separate actuation

DC bus is located on one side of the OEWIG terminals, while the main DC bus of the aircraft electrical power system is connected to the other side of the terminals. In this way, the operation of aircraft main DC power network is prevented from perturbation and disturbance caused by the actuators. The detailed electric circuit of the proposed APU for power generation and management system and the modeling of primary components are presented in the following section.

4.2 Open-end winding induction generator and inverter/rectirﬁer unit model

The electrical circuit diagram of the proposed induction generator based APU for power generation and management system for MEA is shown in Figure 4.3. An open-end winding induction starter/generator is installed on the turbine shaft of an APU. The generator terminals are connected to two inverter/rectiﬁer units (IRUs). The main DC bus is connected to the left side of the generator terminals through IRU-1, while the electric actuation DC bus is located at the right side of the generator terminals through IRU-2. The common mode voltage and current between the two DC buses are inherently eliminated [92]. Hence, all the space vector PWM switching combinations of the IRUs can be utilized.

The voltage equations of the dual inverter/rectiﬁer system are [92, 93]:

dλ v = R i + as (4.1) Aa s as dt

dλ v = R i + bs (4.2) Bb s bs dt

73 APU Turbine Shaft

e a Ls Rs ias A a eb L R ibs To Main s s o Electric DC Bus O B b Actuation e c Ls Rs ics DC Bus C c k c a b d e e

Open-End Winding F

Inverter/Rectifier Unit 1 t Inverter/Rectifier Unit 2 n

Induction Starter/ e r r

Generator u C

DC Voltage Feedback 1 DC Voltage Feedback 2 Controller

Figure 4.3. The proposed power generation and management system architecture with regenerative power absorption capability

dλ v = R i + cs (4.3) Cc s cs dt

where

vAa = vAO − vao + vOo (4.4)

vBb = vBO − vbo + vOo (4.5)

vCc = vCO − vco + vOo (4.6)

In the above equations, vAa, vBb, vCc are the stator phase voltages of the generator; vAO, vBO, vCO are the pole voltages of IRU-1; vao, vbo, vco are the pole voltages of IRU-2; vOo is the voltage diﬀerence between the mid-point of the two separated DC bus. ia, ib, ic and

λas, λbs, λcs are the generator stator currents and ﬂux linkages, respectively. Rs is the stator winding resistance of each phase.

74 By summing up Equations 4.1, 4.2, 4.3, the voltage equation of the open-end winding

induction machine can be written as:

d(λ + λ + λ ) (v + v + v ) − (v + v + v ) + 3v = R (i + i + i ) + as bs cs AO BO CO ao bo co Oo s as bs cs dt (4.7)

Assuming the terminal voltages of IRU-1 and IRU-2 are symmetric, and the induction generator has three phase symmetrical windings, the terminal voltages and generator currents can be expressed as:

vAO + vBO + vCO = 0 (4.8)

vao + vbo + vco = 0 (4.9)

ias + ibs + ics = 0 (4.10)

According to Equations 4.7, 4.8, 4.9, and 4.10, it can be concluded that the mid-points

of the two DC bus O and o are virtually equipotential (i.e., vOo = 0). In rotor ﬂux oriented reference frame, the steady-state generator terminal voltage and

electromagnetic torque can be written as [82]

vqg = vqs1 − vqs2 = Rsiqs + ωeLsids (4.11)

vdg = vds1 − vds2 = Rsids − ωeLsciqs (4.12)

3 T = PP (L − L ) i i (4.13) e 2 s sc qs ds

75 where vqg, vdg, iqs, ids are the q and d axis generator voltages and currents, respectively; vqs1, vds1, vqs2, vds1 are the q and d axis voltages of IRU-1 and IRU-2, respectively. Te is the electromagnetic torque of the generator, while PP is the number of generator pole pairs. Ls, is the generator stator inductance, Lsc is the generator short-circuit inductance. The instantaneous active power delivered to the main DC bus and electric actuation DC bus can be expressed as:

3 p = (v i + v i ) (4.14) dc1 2 qs1 qs ds1 ds

3 p = − (v i + v i ) (4.15) dc2 2 qs2 qs ds2 ds

where pdc1, pdc2 are the active power transmitted to the main DC bus and electric actuation DC bus, respectively. The power generated by the generator is thereby noted as:

3 3 p = p +p = [(v − v ) i + (v − v ) i ] = T ω T ω = ω PP (L − L ) i i g dc1 dc2 2 qs1 qs2 qs ds1 ds2 ds e e , e r 2 r s sc qs ds (4.16)

4.3 System operating principle

The major functionalities of a conventional APU are: i) to start itself using aircraft battery or ground electric power supply (self-start mode); ii) to start the main engine while the aircraft is on the ground (main engine start mode); iii) to provide emergency electrical power (and other types of auxiliary power) during main engine failure (emergency power mode). In addition, a more advanced integrated APU can also act as environmental control system (ECS) and thermal management system (TMS) during the ﬂight mission (cooling mode) [94]. The four major operating modes of the proposed integrated APU for power generation and management system are explained as follow:

76 Electrical Power

APU

APU Turbine Shaft

e a Ls Rs ia e Aircraft battery or uab b Ls Rs ib vab Electric ground power supply C1 vdc1 vdc2 C2 Actuation u e DC Bus bc c Ls Rs ic vbc

Inverter/Rectifier Unit 1 Inverter/Rectifier Unit 2

Open-End Winding Induction Starter/Generator

Figure 4.4. Self-starting mode of operation for the proposed auxiliary power unit

4.3.1 Self-start mode of operation

As shown in Figure 4.4, in self-start mode, the electric actuation DC bus is disconnected from the system. The right side terminals of the induction machine are shorted through IRU- 2. The open-end winding induction generator is thereby transformed into a wye-connected induction motor. IRU-1 is operated as an inverter, driving the induction machine as starter to accelerate the APU turbine shaft using the electrical power from the aircraft battery or ground power supply. Once the APU shaft speed exceeds its self-sustaining speed, the self-start mode is complete.

4.3.2 Main engine start mode of operation

In main engine start mode, the APU engine produces mechanical power to the turbine shaft through combustion of fuel, and drives the induction machines as a generator. As it is illustrated in Figure 4.5, the electric actuation DC bus is still disconnected from the system. The transformed wye-connected induction machine now operating as a generator provides

77 HP Spool HP Spool Jet Fuel Electrical Power Starter/ Turbine Generator APU APU Turbine Shaft

e a Ls Rs ia

u eb i To Main DC ab Ls Rs b vab Electric vdc1 v Distribution C1 dc2 C2 Actuation u e bc c Ls Rs ic vbc DC Bus

Inverter/Rectifier Unit 1 Inverter/Rectifier Unit 2

Open-End Winding Induction Starter/Generator

Figure 4.5. Main engine start mode of operation for the proposed auxiliary power unit electrical power through IRU-1 to the main DC bus. The transmitted power is used by the main engine starter/generator to spin up the HP spool turbine of the main engine. After the main engine is started, the system transits into cooling mode.

4.3.3 Cooling mode of operation

In the cooling mode of operation, the combustor of APU is no longer used. The high stage bleed air from the main engines is ﬁrst cooled down by the APU turbine, and then sends to supply cabin pressurization and air conditioning. Part of the cooled air is also circulated to provide forced air cooling for avionics, ﬂight critical electronics and other liquid cooled heat loads [20].

During a normal ﬂight mission, most of the electrical loads on board are supplied by main engine generation system. Hence, as is shown in Figure 4.6, the main DC bus is disconnected from the APU power management system in the cooling mode. The left side terminals of the open-end winding induction machine are shorted through IRU-1. The APU turbine shaft,

78 Cabin Pressurization High Stage Compressed Bleed Air Cooling Air ECS Equipment APU Main Engine Cooling

APU Turbine Shaft

e a Ls Rs ia

u eb i To Main DC ab Ls Rs b vab Electric vdc1 v Distribution C1 dc2 C2 Actuation u e DC Bus bc c Ls Rs ic vbc

Inverter/Rectifier Unit 1 Inverter/Rectifier Unit 2

Open-End Winding Induction Starter/Generator

Figure 4.6. Cooling mode of operation for the proposed auxiliary power unit powered by the high stage bleed air from main engines, is responsible to supply/absorb the high peak power demand/regeneration from the EHA/EMAs using IRU-2.

4.3.4 Emergency power mode of operation

The emergency mode operation of the proposed system is shown in Figure 4.7. When an in-ﬂight main engine failure occurs, the APU is commanded to operate in full power and produce electricity to support main DC network and potential main engine re-start. In case of dual (all) engine failure, the APU generator is responsible for regulating the main DC bus voltage. As one of the most critical electrical loads on board, the power of EHA/EMAs must be secured at all time during the ﬂight mission. Therefore, in emergency power mode of operation, both sides of the open-end winding induction generator are activated. IRU-2 is responsible for supplying the highly dynamic actuator loads, while IRU-1 is used to provide disturbance-free DC power supply or voltage regulation for the main DC bus.

79 Main DC Electrical Power Jet Fuel Distribution

APU APU Turbine Shaft

e a Ls Rs ia

uab eb L R ib vab Electric To Main DC v s s Distribution C1 dc1 vdc2 C2 Actuation u e bc c Ls Rs ic vbc DC Bus

Inverter/Rectifier Unit 1 Inverter/Rectifier Unit 2

Open-End Winding Induction Starter/Generator

Figure 4.7. Emergency power mode of operation for the proposed auxiliary power unit

4.4 System operating constraints and design considerations

In self-start mode, main engine start mode, and cooling mode of operation, the main DC bus and electric actuation DC bus are isolated because only one side of the open-end winding induction generator is activated. In these modes of operation, the amount of power generation and managed by the proposed system is only a portion of the full system power rating. In the emergency power mode, both sides of the open-end winding induction generator are activated. The proposed power management system is responsible for supplying disturbance-free DC power to the main DC network and regulating the electric actuation DC bus voltage simultaneously. The main DC network has a relatively large but stable load power requirement, whereas the load proﬁle of the actuators, although has smaller peak value, is bidirectional and highly dynamic. The continuous duty of the APU generator can be rated as the sum of nominal main DC load power requirement and average actuator bus load demand. While the actuators are drawing power from the system in the emergency power mode, the open-end winding induction generator is providing electrical power to both

80 ends. In this scenario, the electromagnetic torque command of the generator is increased, but limited by the generators current rating. Since the two IRUs and the generator are connected in series, the current ratings of these units are identical. In other words, the current rating of the proposed system should be designed to fulﬁll the total maximum power demands from both the main DC bus and the actuator DC bus. The maximum current

requirement can be expressed as a function of the APU generator speed ωr and ﬂux current

∗ command ids:

s 2 4 (Pdc1,max + Pdc2,max) ∗ 2 Ismax ≈ ∗ + ids (4.17) 3PP (Ls − Lsc) ωrids As an APU generator, the OEWIG in the proposed system is operated in the flux weakening region for a wide speed range. In most flux weakening algorithms, the flux current

∗ command ids is calculated as a function of the generator speed. Therefore, the current rating of the proposed system can be eventually determined by the APU generator speed range and the maximum power requirement from the main DC bus and actuation DC bus. While the actuators are sending regenerative power back in the emergency operation mode, the power demanded by the main DC network is partially contributed by the regenerative power from the actuators, and the remaining power demand is supplied from the APU shaft power. Since the generator power command is the sum of the power demand of the two DC buses, for the same main DC bus power demand and APU shaft speed, increasing regenerative power from the actuator bus leads to a decreased generator torque command. As the torque command of the generator decreases, the current magnitude falls down. To supply the same amount of active power to the main DC bus with reduced current magnitude, the voltage magnitude of IRU-1 needs to be increased. The larger the regenerative power absorbed by the generator, the higher output voltage is needed on IRU-1 side. As shown in Figure 4.8, in order to deliver maximum active power to the main DC network with limited output voltage, IRU-1 should be commanded to operate under unity power factor.

81 Figure 4.8. Unity power factor operation on main DC network side inverter/rectiﬁer unit

In unity power factor operation, assuming space vector modulation method is used and neglecting the power losses, the active power transmitted through IRU-1 and delivered to the main DC network can be written as:

3 vdc1 pdc1 = m √ |is| (4.18) 2 3

where m is the modulation index of IRU-1, and |is| is the current magnitude of the induction generator. According to (14), the modulation index of IRU-1 would eventually reach its limit with the increasing regenerative power. In this manner, for a given shaft speed, the regenerative power absorption capacity of the proposed power management system can be expressed as:

s ∗ 2 3 ∗ 4 pdc1 ∗ 2 Pdc2 max regen. , ωrPP (Ls − Lsc) ids ∗ 2 − ids (4.19) 2 3 vdc1

82 4.5 Control scheme in emergency power mode

The closed-loop control scheme for the proposed APU for power generation and management system in emergency power mode is shown in Figure 4.9. Two voltage sensors are used to monitor the voltages of two separated DC buses. Two current sensors are used to provide stator current feedback. Neither rotational speed of rotor nor rotor position feedback is essential to control the proposed generation system.

The proposed control scheme is developed based on field oriented control and instantaneous power theory [82, 83]. The field orientation of the proposed control scheme is implemented by an active flux based direct flux observer using inverter terminal voltage and generator stator current feedbacks [88]. The inverter terminal voltages are re-constructed using DC bus voltage feedback and inverter gating signals.

If all of the engines of the aircraft have failed during ﬂight mission, the main DC bus voltage is regulated by the proposed system through PI voltage controller 1. The output of

∗ this controller is the main DC side power command pdc1 . In case of single engine failure,

∗ the main DC bus voltage control loop is disabled, and pdc1 can be provided by main DC bus voltage regulator from the main engine system. To ensure unity power factor operation, the

∗ output voltage vector of IRU-1 is always kept aligned with generator current vector. vds1

∗ and vqs1 can be calculated as:

∗ ids vds1 = v1p (4.20) |is|

∗ iqs vqs1 = v1p (4.21) |is|

∗ where v1p is the voltage magnitude command, which is used as an intermediate control reference to decouple the DC power output of IRU-1 from generator current magnitude variation.

83 Electric Actuation Electric Actuation DC Bus Voltage Regulation DC Bus

vdc2

t i

* - * * * n

V i U dc2 + PI Voltage iq2 qs PI Current vdg

+ * r e Controller 2 + + Controller 1 vds2 i + f - i + Voltage dq t * + SVM c i Comp. + e q1 abc R + * / * * + r

vqs2 e t

ωe Flux Weakening ids + PI Current vqg a r e

Algorithm Controller 2 θf v n

- I ωe OEW a ids IG

2/3PPωe│λd │ ÷ dq ia, ib

t abc i

iqs n

* * * * r e V p v i

dc1 + PI Voltage dc1 1P vds1 f i x dq t Controller 1 ÷ SVM c - x e

abc R

* / v r

qs1 e t │i │ a r

s ids/│is│ θf e

n I vdc1 iqs/│is│

To Main Main DC Bus Voltage Regulation DC Bus

Figure 4.9. Closed-loop control scheme for the proposed APU for power generation and management system in emergency power mode

PI voltage controller 2 is used to regulate the electric actuation DC bus voltage. The gen-

∗ erator q-axis (torque) current command iqs is determined by the adding the q-axis (torque)

∗ current needed for the actuation DC bus iq1 with the q-axis (torque) current demanded by

∗ the main DC bus iq2. Using rotor-ﬂux orientation, two PI current controllers are applied to

regulate ids and iqs.

4.6 Simulation results

A closed-loop simulation for the proposed induction generator based APU for power gener-

ation and management system is conducted in MATLAB/Simulink. In the simulation, a 85

kW rated power (101 kW peak power), 11060 rpm rated speed open-end winding induction

generator is used as the APU generator. The generator is controlled to supply 75 kW 540 V

DC load to the main DC bus and 25 kW 540 V DC load to the separated electric actuation

DC bus at its rated speed.

84 To demonstrate the bidirectional highly dynamic load proﬁle of the EHA/EMAs, the electric actuation DC bus is commanded to toggle between drawing 25 kW power from the APU generator and sending 25 kW regenerative power back with half second intervals from 1.5 s to 3.0 s. The load of the main DC bus is changed from 75 kW to 65 kW at 2.25 s. The main DC bus and electric actuation DC bus voltages of the system is shown in Figure 4.10 and 4.11, respectively. The bidirectional load toggle on the electric actuation bus at 1.5 s, 2.0 s and 2.5 s did not aﬀect the voltage regulation on the main DC bus. The electromagnetic torque and d/q-axis currents characteristics of the APU generator obtained from the simulation are shown in Figure 4.12 and 4.13. The input power of the main DC bus and electric actuation DC bus, as well as the output power of the APU generator are presented in Figure 4.14, 4.15, and 4.16, Clearly, the APU generator successfully supplied/absorbed the highly dynamic power demand/regeneration from the electric actuator bus without disturbing the power delivery to the main DC bus.

545

544

543 )

V 542 (

e g

a 541 t l o v

s 540 u b

C 539 D

n i

a 538 M 537

536

535 1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 3 Time (s)

Figure 4.10. The main DC bus voltage characteristics of the proposed APU for power generation and management system

85 560

) 555 V (

e g

a 550 t l o v

s 545 u b

C D

540 n o i t a 535 u t c A

i 530 r t c e l

E 525

520 1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 3 Time (s)

Figure 4.11. The electric actuation DC bus voltage characteristics of the proposed APU for power generation and management system

-30

-40 ) m N ( -50 e u q r o t

-60 c i t e n

g -70 a m o r t

c -80 e l E

-90

-100 1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 3 Time (s)

Figure 4.12. The APU generator electromagnetic torque characteristics of the proposed APU for power generation and management system

86 0.5

) d-axis current . u

. q-axis current -0.5 p (

t n e r r

u -1 c

s i x a -

q -1.5 , D

-2

-2.5 1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 3 Time (s)

Figure 4.13. The APU generator d and q axis current characteristics of the proposed APU for power generation and management system

4 x 10 -6.4

-6.6 ) W (

e -6.8 w o p

t u

p -7 n i

s u b

C -7.2 D

n i a

M -7.4

-7.6 1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 3 Time (s)

Figure 4.14. The main DC bus input power characteristics of the proposed APU for power generation and management system

87 4 x 10 4 )

W 3 (

r e w

o 2 p

t u

p 1 n i

s u b 0 C D

n o

i -1 t a u t

c -2 A

c i r t

c -3 e l E

-4 1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 3 Time (s)

Figure 4.15. The electric actuation DC bus input power characteristics of the proposed APU for power generation and management system

4 x 10 -3

-4 ) W ( -5 r e w o

p -6

t u p t

u -7 o

r o t

a -8 r e n e

g -9

U P -10A

-11 1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 3 Time (s)

Figure 4.16. The APU generator output power characteristics of the proposed APU for power generation and management system

88 4.7 Summary

In this chapter, an induction generator based auxiliary power unit for power generation and management system for MEA is presented. The proposed APU power generation and management system provides direct power flow path from the EHA/EMAs to the power source, but prevents the main DC network from perturbation and disturbance caused by the actuators. A separate DC bus is created for the EHA/EMAs without adding significant hardware installment. Both main and electric actuation DC bus voltages of the system can be well-regulated simultaneously. The feasibility of operation of the proposed system is verified in MATLAB/Simulink and the results are presented.

89 CHAPTER 5

A MODULAR BACK-UP POWER LINK UNIT FOR A

RE-CONFIGURABLE FAULT-TOLERANT ACTUATION SYSTEM

ARCHITECTURE WITH SEPARATED POWER SUPPLY IN MORE

ELECTRIC AIRCRAFT

As mentioned in the previous chapters, in MEA system with EHA/EMAs, separate electric actuator bus can protect the main electrical power network from disturbance caused by the

EHA/EMAs. However, if an equipment failure occurs in the primary power link (such as the new APU proposed in Chapter 4) between aircraft main power grid and separated actuation power bus, the flight safety critical actuator loads are isolated from the rest of the aircraft electrical power network. Back-up power links between the separated actuator loads and the electrical power source(s) can reduce the requirement of the aircraft emergency batteries and improve the system reliability and availability. In this chapter, a modular back-up power link unit for a re-configurable fault-tolerant actuation system architecture with separated power supply in more electric aircraft is proposed. In the proposed architecture, each individual actuation drive unit is configured as three-phase H-bridge converters in normal operation. In case of equipment failure occurs in the primary power link between aircraft main power grid and separated actuation power bus, the redundant/idling actuation drive units can be transformed into modular back-up power link units with two six-switch two level inverter/rectifiers. One of the inverter/rectifiers is connected with the main DC networks, whereas the separated actuator bus is connected with the other inverter/rectifier. In this way, multiple back-up power links for the separated actuator loads can be created without adding significant hardware. A closed-loop control scheme is developed so that constant power can be drew from the main DC networks through the re-configured back-up power link units.

90 5.1 Introduction

In aircraft system, the flight control system is one of the utmost critical systems which responsible for governing the motion of the aircraft. In conventional aircraft, multiple redundant architectures for aircraft hydraulic and electrical system are used to secure integrity and reliability of the flight control system. A simplified block schematic diagram of a conventional hydraulic actuator with multiple redundancy is shown in Figure 5.1[95], in the conventional centralized hydraulic actuation system, four identical lanes of actuation control and two channels of hydraulic power inputs are available for one single actuator.

Hydraulic Hydraulic System I System II Electrical Electrical Control Control Signals Signals

Sol I Sol II

Control Electrical Servo Valves Control Hydraulic Signaling Valves Hydraulic Signaling Actuator Signals x4 x4 Servo Valves Position Feedback RAM Position Feedback

Figure 5.1. Simpliﬁed block schematic diagram of a conventional hydraulic actuator with multiple redundancy [95]

In the trend towards MEA, the traditional hydraulic ﬂight control actuators with centralized hydraulic system are being replaced by EHAs or EMAs. Compared to the traditional actuators, EHAs and EMAs have lower weight and volume, increased reliability and reduced maintenance [59, 60, 61]. As one of the most recent MEA, Airbus A380 uses electrical back- up hydraulic actuators (EBHAs) to power the spoiler surfaces and rudder sections. As is shown in Figure 5.2 [58], the EBHAs combines the features of hydraulic actuators and EHAs. In EBHAs, the hydraulic powered servovalves is used in normal operation, whereas the electrically power motor pump is used to provide back-up actuation. Similar to the conventional

91 Electrical Electronics system (power) Mode Motor selector device Hydraulic system (power)

Ram

Figure 5.2. Electrical backup hydraulic actuator used on Airbus A380 [58] hydraulic actuators, the EBHAs also have multiple redundancies in terms of actuation power lanes and control channels. In A380, EBHAs can receive hydraulic power inputs from two independent centralized hydraulic system, and electrical power supply from three different AC buses [58]. With continuous advancement in aircraft hydraulic system electrification, the centralized hydraulic system is predicted to be completely replaced by new flight control actuation systems using EHAs and EMAs. However, the EHA/EMAs have highly dynamic power profile, which includes high peak power demand and power regeneration [65, 66, 67]. Since power regeneration into electrical power grid is prohibited by current regulations [68], the electrical separation between and actuator loads and main electric power network becomes necessary. The separated electric power bus for actuators can be created by either additional power electronic devices with electrical storage elements (ESEs) [22] or new electrical power generation and management architectures (i.e., the new APU proposed in Chapter 4 )[96]. However, in case of equipment failure occurs in the power link between the separated actuator bus and main electric power network, the actuators, as one of the utmost safety critical electrical loads on board, are isolated from the most of the available power sources in the aircraft. Although aircraft batteries can provide emergency power to the actuators, to fulfill the ETOPS requirement and reduce the reserved aircraft battery capacity, redundant power

92 Inverter/ converter unit

Isolated ES/G Power Electric drives Auxiliary power Converter EHA/EMA unit (APU) with ESEs Redundant power path needed Power ES/G Distribution Other Electric Loads Network Engine Inverter/ converter unit (a) MEA electrical power system architecture using power converters and ESEs for actuator load separation Inverter/ converter unit

Electric drives EHA/EMA Inverter/ converter unit

ES/G Redundant power path needed Auxiliary power unit (APU)

Power ES/G Distribution Other Electric Loads Network Engine Inverter/ converter unit (b) MEA electrical power system architecture using APU for actuator load separation

Figure 5.3. The demand of redundant power path for diﬀerent type of electrical power system with separated actuator buses

ﬂow path to the primary or secondary electrical power sources should be available for the separated actuation bus. The demand of redundant power path for diﬀerent type of electrical power system with separated actuator buses is shown in Figure 5.3.

Multiple redundant power path or redundant separated buses can be provided by adding extra hardware (such as back-up DC/DC converters with ESEs) to the new MEA architecture using EHA/EMAs. Nonetheless, considering the excellent re-conﬁgurability of electrical power system, it is possible to take advantages of the redundant/idling actuator drive units

93 and create back-up power path for the separated actuator bus without adding significant hardware. For this purpose, new re-configurable fault-tolerant flight control actuation system architecture for the separated power supply is expected.

5.2 A re-conﬁgurable fault-tolerant actuation system architecture with modular back-up power links

The configuration of the proposed re-configurable fault-tolerant actuation system architecture with modular back-up power links (MBPLs) for separated actuator loads in MEA is shown in Figure 5.4. In the proposed architecture, the electric actuation DC bus is separated from the main DC distribution by the primary power link, which can be referred to an isolated power converter with ESEs [22, 21], or an open-end winding induction generator based APU generation system [96]. An emergency power battery is equipped on the separated electric actuation DC bus as reserved energy storage device. The proposed MBPL unit con- sist of an open-end winding interior permanent magnet synchronous motor (IPMSM), two inverter/rectifier units (IRUs), and a double pole double throw (DPDT) power switch. The DPDT power switch can shift the proposed MBPL unit between normal operation mode and emergency charging operation mode. The normal mode (NM) of operation of the proposed system is shown in Figure 5.5. In this mode of operation, the primary power link is responsible for supplying the highly dynamic actuator loads on the separated bus while drawing/sending constant power flow from/to the main DC distribution. The MBPL units operate as actuation motor drive systems, both sides of the IRUs of the MBPL units are connected to the electric actuation DC bus. The emergency power battery is not activated in the normal mode of operation. In case of equipment failure in the primary power link, the proposed system is transformed into the emergency charging mode (ECM) of operation. As is shown in Figure 5.6, in emergency charging mode, the emergency power battery is activated. Constant DC power

94 Electric To Main DC Primary Power Link for Load Separation Actuation Distribution C1 C2 DC Bus

Emergency Power Battery Actuator shaft 1

e a Ls Rs

eb Ls Rs

ec Ls Rs

Inverter/Rectifier Inverter/Rectifier Unit 1(a) IPM Motor 1 Unit 1(b)

Modular Back-up Power Links

Actuator shaft n

e a Ls Rs

eb Ls Rs

ec Ls Rs

Inverter/Rectifier Inverter/Rectifier Unit n(a) IPM Motor n Unit n(b)

Figure 5.4. The proposed re-conﬁgurable fault-tolerant actuation system architecture with modular back-up power links for separated actuator loads in MEA

ﬂows from the main DC distribution network are transmitted to the separated actuation

DC bus through “qualiﬁed” MBPL units. The “qualiﬁed” MBPL units refer to the redundant/idling actuator drive units while the fault condition occurs in primary power link. By connecting one side of their IRUs to the main DC bus, the back-up power links for charging the emergency power battery from the main DC distribution can be established.

During the entire ﬂight mission, available idling actuator drive units for back-up charging operation include:

95 Electric To Main DC Primary Power Link for Load Separation Actuation Distribution C1 C2 DC Bus

Emergency Power Battery Actuator shaft 1

e a Ls Rs

eb Ls Rs

ec Ls Rs

Inverter/Rectifier Inverter/Rectifier Unit 1(a) IPM Motor 1 Unit 1(b)

Actuator shaft n

e a Ls Rs

eb Ls Rs

ec Ls Rs

Inverter/Rectifier Inverter/Rectifier Unit n(a) IPM Motor n Unit n(b)

Figure 5.5. Normal mode of operation of the proposed re-conﬁgurable fault-tolerant actuation system architecture

• Back-up actuator drive units in the actuation system, available as long as their primary

counterparts work in non-faulty condition;

• Slow response secondary ﬂight control actuator drive units such as ﬂap and slat actu-

ator drives, available when no high lift control is required;

• Landing gear actuator drive units, generally available during the entire ﬂight mission

expect landing phase.

96 Electric To Main DC Primary Power Link for Load Separation Actuation Distribution C1 C2 DC Bus

Actuator shaft 1 (Idling Actuator) Emergency Power Battery

e a Ls Rs

eb Ls Rs

ec Ls Rs

Inverter/Rectifier Inverter/Rectifier Unit 1(a) IPM Motor 1 Unit 1(b)

Actuator shaft n (Activated Actuator)

e a Ls Rs

eb Ls Rs

ec Ls Rs

Inverter/Rectifier Inverter/Rectifier Unit n(a) IPM Motor n Unit n(b)

Figure 5.6. Emergency charging mode of operation of the proposed re-conﬁgurable fault- tolerant actuation system architecture

In ECM, all of the above actuator drive units are the candidates of the proposed MBPL units. The electric power management computer in the MEA is responsible for the de- ployments of the MBPL units. Only the redundant/idling actuators will be commanded to provide back-up charging to the separated actuator loads. In the proposed re-conﬁgurable fault-tolerant actuation system, actuation commands from the ﬂight control system always has higher priority to the back-up charging commands. If an actuation command is assigned to an idling actuator during back-up charging, the charging operation will be halted. The

97 Electric A o a Actuation DC Bus

Actuator shaft B s R e b b

s L L s

a IPM e R Motor s e c Ls Rs

C c

Figure 5.7. Normal mode of operation of the proposed modular back-up power link unit assigned MBPL unit will be re-conﬁgured back to the actuator drive unit to execute the actuation command. Detailed modeling and operation of individual MBPL unit will be discussed in the following section.

5.3 Modeling and operation of the modular back-up power link unit

The electrical circuit diagram of the proposed MBPL unit in NM is shown in Figure 5.7. In this mode of operation, the two IRUs are connected to the same separated DC bus. The open-end winding IPMSM drive system can be recognized as three independent H-bridge drives. This topology is widely used to improve reliability and fault tolerance of the drive system in aircraft actuation applications [33]. The operation and control algorithm of such multiple independent single-phase drives is well recorded in the literature [37, 97, 98].

98 Actuator shaft

e a Ls Rs A a eb L R To Main s s o Electric DC Bus O B b Actuation e c Ls Rs DC Bus C c

Inverter/Rectifier Unit (a) Open-End Winding Inverter/Rectifier Unit (b) IPM Motor

Figure 5.8. Emergency charging mode of operation of the proposed modular back-up power link unit

If the actuator drive unit in Figure 5.7 is idling when equipment failure occurs in the primary power link, the entire actuator drive unit is transformed into a back-up power link unit as shown in Figure 5.8. In this mode of operation, the main DC bus is connected to the left side of the IPMSM terminals through IRU (a), while the electric actuation DC bus is connected to the right side of the motor terminals through IRU (b). The common mode voltage and current between the two DC buses are inherently eliminated [92]. Consequently, all the space vector PWM switching combinations of the IRUs can be utilized. The voltage equations of the IPMSM with dual inverter/rectiﬁer drive system are [92, 93]:

dλ v = R i + as (5.1) Aa s as dt

dλ v = R i + bs (5.2) Bb s bs dt

dλ v = R i + cs (5.3) Cc s cs dt where

vAa = vAO − vao + vOo (5.4)

99 vBb = vBO − vbo + vOo (5.5)

vCc = vCO − vco + vOo (5.6)

In the above equations, vAa, vBb, vCc are the stator phase voltages of the IPMSM; vAO, vBO, vCO are the pole voltages of IRU (a); vao, vbo, vco are the pole voltages of IRU (b); vOo is the voltage diﬀerence between the mid-point of the two separated DC bus. ia, ib, ic and

λas, λbs, λcs are the motor stator currents and ﬂux linkages, respectively. Rs is the stator winding resistance of each phase. By summing up Equations 5.1, 5.2, 5.3, the voltage equation of the open-end winding IPMSM can be written as:

d(λ + λ + λ ) (v + v + v ) − (v + v + v ) + 3v = R (i + i + i ) + as bs cs AO BO CO ao bo co Oo s as bs cs dt (5.7) Assuming the terminal voltages of IRU (a) and IRU (b) are symmetric, and the IPMSM has three phase symmetrical windings, the terminal voltages and motor currents can be expressed as:

vAO + vBO + vCO = 0 (5.8)

vao + vbo + vco = 0 (5.9)

ias + ibs + ics = 0 (5.10)

According to Equations 5.7, 5.8, 5.9, and 5.10, it can be concluded that the mid-points

of the two DC bus O and o are virtually equipotential (i.e., vOo = 0).

100 In rotor ﬂux oriented reference frame, the IPMSM terminal voltage and electromagnetic torque can be written as [82]

di v = v − v = R i + L q + ω L i + ω φ (5.11) qm q(a) q(b) s q q dt e d d e m

di v = v − v = R i + L d − ω L i (5.12) dm d(a) d(b) s d d dt e q q

3 T = PP [φ i − (L − L ) i i ] (5.13) e 2 m q q d q d

where vqm, vdm, iq, id are the q and d axis IPMSM voltages and currents, respectively; vq(a), vd(a), vq(b), vd(b) are the q and d axis voltages of IRU (a) and (b), respectively. Te is

the electromagnetic torque of the IPMSM, while PP is the number of motor pole pairs. Lq

and Ld are the motor stator q and d axis inductance. φm is the ﬂux linkage created by the permanent magnets in the rotor. The instantaneous active power delivered to the IPMSM from the main DC bus and electric actuation DC bus can be expressed as:

3 p = v i + v i (5.14) dc(a) 2 q(a) q d(a) d

3 p = − v i + v i (5.15) dc(b) 2 q(b) q d(b) d

where pdc(a), pdc(b) are the active power transmitted to the IPMSM from the main DC bus and electric actuation DC bus, respectively. The power consumed by the IPMSM is thereby noted as:

3 3 p = p +p = v − v i + v − v i ω PP [φ i − (L − L ) i i ] m dc(a) dc(b) 2 q(a) q(b) q d(a) d(b) d , 2 r m q q d q d (5.16)

101 id Rs Ld

v * + + v * d(a) - - d(b)

Figure 5.9. Equivalent circuit of the proposed modular back-up power link unit in emergency charging mode

Since the emergency charging operation takes place while the actuator is idling, the actuator motor shaft has to be kept as standstill and no electromagnetic torque should be produced during the emergency charging operation. According to the torque equation of the IPMSM (Equation 5.13), the q axis current of the motor should be regulated as zero throughout the whole charging process (i.e., iq = 0). As a result, only d axis current of the motor drive system can be utilized to charge the emergency battery. The voltage equation of the IPMSM at standstill (i.e., ωe = 0) with no q axis current can be written as:

vqm = 0 (5.17)

di v = v − v = R i + L d (5.18) dm d(a) d(b) s d d dt The polarity of the d axis current during ECM can be either positive or negative. If a negative d axis current is utilized, the charging current through an individual IPMSM drive unit should be limited to prevent the permanent magnet (PM) from demagnetization. In case of positive d axis current charging, the core of the IPMSM may be saturated, resulting in slight decrease in the d axis inductance; however, the positive d axis current would not

102 demagnetize the PM, resulting in a much wider operating range of charging current. In the proposed system, positive d axis current charging is selected. Negating the losses in IRUs and potential saturation eﬀect in IPMSM, the equivalent circuit of an individual MBPL unit in ECM is shown in Figure 5.9. In most MEA electrical power system, the voltages on main DC bus and actuation DC bus are identical [56]. In order to transmit electric power from the (a) side to the (b) side of the MBPL unit, the

∗ commanded (a) side IRU output voltage vd(a) should always be larger than the commanded ∗ (b) side IRU output voltage vd(b). To maximize the charging power for a given charging current of the MBPL unit, the IRU (a) output voltage can be commanded as its maximum value for the whole charging operation.

5.4 Closed-loop control scheme of the modular back-up power link unit for emergency charging

The closed-loop control scheme for the proposed MBPL unit in ECM is shown in Figure 5.10. In this control scheme, two PI controllers are used to regulate the d and q axis current of the IPMSM, two current sensors are used to provide stator current feedback. Since multiple MBPL units are available in ECM, the charging current of each MBPL unit is designed as a control input, which can be commanded by the battery management system (BMS). The control scheme is operated under rotor ﬂux orientation [82]. Rotor position feedback can be obtained by the position senor installed on the actuator motor shaft. The main DC bus side IRU output voltages are commanded to be constant, which can be expressed as:

Vdcm vd(a) = √ (5.19) 3

vq(a) = 0 (5.20)

103 Electric Actuation DC Bus

* * r

v e iq =0 PI Current qm - i f

* i t

Controller 1 v c + d(b) )

+ e - b

dq ( R

t SVM / i r n abc e t r * U

* * vq(b) e id + PI Current vdm - v n I Controller 2 θf - + IPMSM id dq ia, ib

abc r

iq e i f i t c * ) e a (

vq(a) =0 dq R t /

SVM i * r n e

vd(a) = abc t r U e v n I θf

To Main DC Bus

Figure 5.10. Closed-loop control scheme for the proposed modular back-up power link unit in emergency charging mode

where Vdcm is the main DC bus voltage. The voltage commands of IRU (a) are used to calculate the voltage commands of IRU (b) as:

vq(b) = vq(a) − vqm (5.21)

vd(b) = vd(a) − vdm (5.22)

5.5 Simulation results

A closed-loop simulation for the proposed MBPL unit for a re-conﬁgurable fault-tolerant actuation system architecture is conducted in MATLAB/Simulink. In the simulation, a

104 5.5 kW, 2000 rpm open-end winding IPMSM is used as the actuator motor. The motor is controlled to transmit 5 kW power from the 270 V main DC bus to the 270 V electric actuation DC bus at standstill. A 270 V emergency power battery in connected to the separated actuation DC bus. The charging current is commanded to change from 18.5 A to

10 A at 0.1 s.

The d/q-axis currents characteristics of the proposed MBPL unit is shown in Figure 5.11.

The d/q-axis output voltage characteristics of the two IRUs obtained from the simulation are shown in Figure 5.12 and 5.13, respectively. The three phase currents of the IPMSM, as well as the emergency battery voltage and charging current characteristics are presented in

Figure 5.14, and 5.15. Clearly, the MBPL unit successfully supplied constant commended current to the separated actuator bus.

) id

A 15 (

t n e r r u

c 10 id

s i x a - q ,

d 5 * iq and iq

-5 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Time (s)

Figure 5.11. The d and q axis current characteristics of the proposed modular back-up power unit for re-conﬁgurable fault-tolerant actuation system architecture

105 160

159

158 * Vd(a) 157 ) V (

e 156 g a t l

o 155 v

s i x

a 154 - d * 153 Vd(b)

152

151

150 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Time (s)

Figure 5.12. The d axis voltage characteristics of the proposed modular back-up power unit for re-conﬁgurable fault-tolerant actuation system architecture

0.01

0.008

0.006

0.004 ) V (

e 0.002 g a t l

o 0 v

s i x

a -0.002 -

q * * -0.004 Vq(a) and Vq(b)

-0.006

-0.008

-0.01 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Time (s)

Figure 5.13. The q axis voltage characteristics of the proposed modular back-up power unit for re-conﬁgurable fault-tolerant actuation system architecture

106 25

) 15 A (

t n e

r 10 r u c

, i 5 a B , A

e s

a 0 h P ib and ic -5

-10

-15 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Time (s)

Figure 5.14. The three phase current characteristics of the proposed modular back-up power unit for re-conﬁgurable fault-tolerant actuation system architecture

300 ) A (

t 250 n e r r Emergency power u c

g 200 battery voltage n i g r a h c

d 150 n a

) V (

g 100 a t l Battery charging o v current y r 50 e t t a B

0 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Time (s)

Figure 5.15. The emergency power battery voltage and current characteristics of the proposed modular back-up power unit for re-conﬁgurable fault-tolerant actuation system architecture

107 5.6 Summary

In this chapter, a modular back-up power link unit for a re-configurable fault-tolerant actuation system architecture with separated power supply in more electric aircraft is proposed. In the proposed architecture, the redundant/idling actuator drive units are re-configured and utilized to provide back-up power path between the main electric power grid and separated actuation power bus of the MEA without adding significant hardware. The proposed MBPL unit and re-configurable fault-tolerant actuation system architecture can be used to improve the system reliability and availability of the APU based power management and generation system proposed in Chapter 4. The configuration and operating principle of the proposed re-configurable fault-tolerant actuation system are presented. Modeling and operation of the proposed modular back-up power link units are discussed. A closed-loop control scheme for extracting constant and steady power flow from the primary power source through the modular back-up power link unit is developed. The feasibility of the operation of proposed modular back-up power link unit is demonstrated by simulation in MATLAB/Simulink.

108 CHAPTER 6

CONCLUSION AND FUTURE WORK

This dissertation presents new electrical power generation and distribution architectures for the main engine, the auxiliary power unit and the actuation system in more electric aircraft.

Induction generator with open-end winding topology is used as the main starter/generator in both the main engine generation architecture and the auxiliary power unit for regenerative power management system. The application of induction generator in the proposed systems improves the system power density compared to synchronous generator based systems, and avoids the excessive faulty current issue caused by PM generators. A modular back-up power link unit for re-conﬁgurable fault-tolerant actuation system architecture is also proposed to provide additional power supply path for the ﬂight safety critical actuator loads in the proposed regenerative power management system with separated actuation bus.

The above more electric architectures can be combined as a new electrical power generation and distribution system for commercial transport aircraft.

6.1 Conclusions

In this dissertation, more electric architectures in three necessary development areas for transforming the conventional commercial transport aircraft toward the more electric aircraft are investigated.

Firstly, the main engine electrical power generation system architectures for present and future MEA electrical power system with diﬀerent electric starter/generator candidates are studied. An induction generator based AC/DC hybrid electric power generation system for

MEA is proposed. Because of the self-excited capability of induction generator, the proposed

AC/DC hybrid generation architecture can supply constant voltage variable frequency power directly from one side of the generator winding terminals like an AC primary generation

109 system. On the other side of the generator winding terminals, the more advanced paralleled

DC bus power generation architecture from DC primary generation system is also reserved.

As a result, the overall hardware requirement of the proposed system is reduced compared to both AC and DC primary generation systems. Control schemes based on instantaneous power control theory and ﬁeld orientation control theory are developed to regulate the AC load voltage and coordinate DC power generation between the two generators. Both AC and DC output voltages of the system can be well-regulated with generator speed, AC and

DC side load and DC power output command variation. The feasibility of operation of the proposed system is demonstrated by both computer simulation and hardware-in-the-loop real-time emulation.

Secondly, the architectural solutions for managing the high peak and regenerative power

ﬂow from the electro-hydrostatic actuators and/or electro-mechanical actuators are investigated. An auxiliary power unit for power generation and management system that allows the regenerative power from the actuators to be absorbed by the turbine shaft of the APU is proposed. An open-end winding induction starter/generator is used to provide direct power

flow path from the EHA/EMAs to the power source, and to create a separate electric actuation bus without significant additional hardware requirement. In this way, the operation of aircraft main DC power network is prevented from perturbation and disturbance caused by the actuators. The configuration and modeling of the proposed induction generator based

APU for power generation and management system are presented. System operation for all operating conditions of the aircraft is analyzed. A closed-loop control scheme for regulating both main DC bus and actuation DC bus voltages in aircraft emergency power mode is developed based on FOC and instantaneous power theory. The feasibility of operation of the proposed system is demonstrated by simulation in MATLAB/Simulink.

Finally, the possibility of enhancing the reliability and availability of the ﬂight safety critical actuation system and the proposed APU based regenerative power management system

110 without additional significant hardware is explored. A modular back-up power link unit for re-configurable fault-tolerant actuation system architecture is proposed. This architecture provides back-up power path between the main electric power grid and separated actuation power bus of the MEA without adding significant hardware. In the proposed architecture, each individual actuator motor drive is configured as three-phase H-bridge converter in normal operation. In case of the failure occurs in the path towards primary power source of the actuation bus, the three-phase H-bridge converter can be transformed into two six-switch two level inverter/rectifier. One of the inverter/rectifier is connected with the electrical bus of primary power source, whereas the separated actuator bus is connected with the other inverter/rectifier. In this way, each of the actuator motor drives can act as a back-up power link between the main source and the separated bus. Modeling and operation of the proposed modular back-up power link units are discussed. A closed-loop control scheme for extracting constant and steady power flow from the primary power source through the modular back-up power link unit is developed. The feasibility of the operation of proposed modular back-up power link unit is demonstrated by simulation in MATLAB/Simulink. The proposed more electric architectures in this dissertation provide solutions for electrification development of aircraft systems in terms of enhancing the electric power generation capacity of the aircraft, reducing the hardware requirement of the electric power generation and distribution system, managing the high peak and regenerative power flow from the electro-hydrostatic actuators and/or electro-mechanical actuators, and enhancing the reliability and availability of the flight safety critical actuation system and the regenerative power management system.

6.2 Future work

The work presented in this dissertation proposes architectural designs for the electrical power generation and distribution system, regenerative power management system, and

111 re-configurable fault-tolerant actuation system. Further research in the scope of this dissertation includes further expansion and evaluation of the above individual architectures, investigation of the transition between different mode of operation (including various faulty operating conditions) of the proposed systems, and system level combinations and integration of the above systems. In the electrical power generation and distribution system, extended load sharing strategies between multiple engines and/or APU(s) (including system operation under single and dual engine failures) can be investigated. The electrical power quality at the moment of fault is to be examined. Additional protection strategies may also be considered. In the APU based regenerative power management system, the simulation results presented in the dissertation demonstrated the feasibility of the proposed control scheme. Further development of a hardware experimental verification would evaluate and validate the system performance with practical actuator load profiles. In re-configurable fault-tolerant actuation system, further investigation in system level battery charging management strategies using multiple modular back-up power link units can enable the system integration of the actuation system and APU based regenerative power management system. Fast response transformation between normal and emergency charging mode of operations may also be investigated.

112 REFERENCES

[1] International Civil Aviation Organization, Global air transport outlook to 2030 and trends to 2040. International Civil Aviation Organization, 2013.

[2] Internatoinal Civil Aviation Organization, “Climate Change : Programme of Action,” 2009.

[3] European Commission, “Flightpath 2050 Europe’s Vision for Aviation,” tech. rep., 2011.

[4] K. Emadi and M. Ehsani, “Aircraft power systems: technology, state of the art, and future trends,” IEEE Aerospace and Electronic Systems Magazine, vol. 15, no. 1, pp. 28– 32, 2000.

[5] J. Rosero, J. Ortega, E. Aldabas, and L. Romeral, “Moving towards a more electric aircraft,” IEEE Aerospace and Electronic Systems Magazine, vol. 22, pp. 3–9, mar 2007.

[6] S. Cutts, “A collaborative approach to the More Electric Aircraft,” in International Conference on Power Electronics Machines and Drives, vol. 2002, pp. 223–228, IEE, 2002.

[7] B. Sarlioglu and C. T. Morris, “More Electric Aircraft: Review, Challenges, and Op- portunities for Commercial Transport Aircraft,” IEEE Transactions on Transportation Electriﬁcation, vol. 1, pp. 54–64, jun 2015.

[8] H. Abu-Rub, M. Malinowski, and K. Al-Haddad, Power Electronics for Renewable En- ergy Systems, Transportation and Industrial Applications. John Wiley & Sons, 2014.

[9] B. S. Bhangu and K. Rajashekara, “Control strategy for electric starter generators embedded in gas turbine engine for aerospace applications,” in IEEE Energy Conversion Congress and Exposition: Energy Conversion Innovation for a Clean Energy Future, ECCE 2011, Proceedings, pp. 1461–1467, 2011.

[10] K. Muehlbauer and D. Gerling, “Two-Generator-Concepts for Electric Power Gener- ation in More Electric Aircraft Engine,” XIX International Conference on Electrical Machines, p. 5, 2010.

[11] M. J. Cullen, “Permanent Magnet Generator Options for the More Electric Aircraft,” in Power Electronics, Machines and Drives, 2012.

[12] W. U. N. Fernando, M. Barnes, and O. Marjanovic, “Direct drive permanent magnet generator fed AC-DC active rectiﬁcation and control for more-electric aircraft engines,” IET Electric Power Applications, vol. 5, no. 1, pp. 14–27, 2011.

113 [13] F. Khatounian, E. Monmasson, F. Berthereau, E. Delaleau, and J. Louis, “Control of a doubly fed induction generator for aircraft application,” in IECON’03. 29th Annual Conference of the IEEE Industrial Electronics Society (IEEE Cat. No.03CH37468), vol. 3, pp. 2711–2716, 2003.

[14] T. Lipo and I. Alan, “Starter/generator employing resonant-converter-fed induction machine. I. Analysis,” IEEE Transactions on Aerospace and Electronic Systems, vol. 36, no. 4, pp. 1309–1328, 2000.

[15] R. Bojoi, A. Cavagnino, A. Tenconi, and S. Vaschetto, “Control of Shaft-Line-Embedded Multiphase Starter/Generator for Aero-Engine,” IEEE Transactions on Industrial Elec- tronics, vol. 63, pp. 641–652, jan 2016.

[16] M. Sinnett, “787 No-Bleed Systems: saving fuel and enhancing operational eﬃciencies,” Boeing Aero Magazine, pp. 6–11, 2007.

[17] J. Brombach, A. Lucken, B. Nya, M. Johannsen, and D. Schulz, “Comparison of diﬀerent electrical HVDC-architectures for aircraft application,” in 2012 Electrical Systems for Aircraft, Railway and Ship Propulsion, pp. 1–6, IEEE, oct 2012.

[18] K. Rajashekara, J. Grieve, and D. Daggett, “Hybrid fuel cell power in aircraft,” IEEE Industry Applications Magazine, vol. 14, pp. 54–60, jul 2008.

[19] A. Lucken, J. Brombach, and D. Schulz, “Design and protection of a high voltage DC onboard grid with integrated fuel cell system on more electric aircraft,” in Electrical Systems for Aircraft, Railway and Ship Propulsion, pp. 1–6, IEEE, oct 2010.

[20] J. Oliver, P. Zumel, M. Sanz, C. Raga, D. Izquierdo, O. Garcia, A. Barrado, R. Prieto, R. Azcona, B. Delicado, and J. Cobos, “High level decision methodology for the selection of a fuel cell based power distribution architecture for an aircraft application,” in 2009 IEEE Energy Conversion Congress and Exposition, pp. 459–464, IEEE, sep 2009.

[21] Y. Deng, S. Y. Foo, and I. Bhattacharya, “Regenerative electric power for More Electric Aircraft,” in IEEE SOUTHEASTCON 2014, pp. 1–5, IEEE, mar 2014.

[22] P. Wheeler, A. Trentin, S. Bozhko, and J. Clare, “Regeneration of energy onto an aircraft electrical power system from an electro-mechanical actuator,” in 2012 Electrical Systems for Aircraft, Railway and Ship Propulsion, pp. 1–6, IEEE, oct 2012.

[23] L. Faleiro, “Beyond the More Electric Aircraft,” Aerospace America, pp. 35–40, 2005.

[24] P. Wheeler and S. Bozhko, “The More Electric Aircraft: Technology and challenges.,” IEEE Electriﬁcation Magazine, vol. 2, pp. 6–12, dec 2014.

114 [25] C. Spitzer, “The All-Electric Aircraft: A Systems View and Proposed NASA Research Programs,” IEEE Transactions on Aerospace and Electronic Systems, vol. AES-20, pp. 261–266, may 1984.

[26] M. J. Cronin, A. P. Hays, F. B. Green, N. A. Radovcich, C. W. Helsley, and W. L. Rutchik, “Integrated digital/electric aircraft concepts study,” tech. rep., 1985.

[27] M. Hirst, A. McLoughlin, P. Norman, and S. Galloway, “Demonstrating the more electric engine: a step towards the power optimised aircraft,” IET Electric Power Applications, vol. 5, no. 1, p. 3, 2011.

[28] R. Jones, “The More Electric Aircraft: the past and the future?,” in IEE Colloquium. Electrical Machines and Systems for the More Electric Aircraft, vol. 1999, pp. 1–1, IEE, 1999.

[29] M. Provost, “The More Electric Aero-engine: a general overview from an engine man- ufacturer,” in International Conference on Power Electronics Machines and Drives, vol. 2002, pp. 246–251, IET, 2002.

[30] G. Raimondi, T. Sawata, M. Holme, A. Barton, G. White, J. Coles, P. H. Mellor, and N. Sidell, “Aircraft embedded generation systems,” in International Conference on Power Electronics Machines and Drives, vol. 2002, pp. 217–222, IET, 2002.

[31] A. Cavagnino, Z. Li, A. Tenconi, and S. Vaschetto, “Integrated Generator for More Electric Engine: Design and Testing of a Scaled-Size Prototype,” IEEE Transactions on Industry Applications, vol. 49, pp. 2034–2043, sep 2013.

[32] J. Wang, K. Atallah, and D. Howe, “Optimal Torque Control of Fault-Tolerant Per- manent Magnet Brushless Machines,” in IEEE Transactions on Magnetics, vol. 39, pp. 2962–2964, 2003.

[33] W. Wenping Cao, B. C. Mecrow, G. J. Atkinson, J. W. Bennett, and D. J. Atkinson, “Overview of Electric Motor Technologies Used for More Electric Aircraft (MEA),” IEEE Transactions on Industrial Electronics, vol. 59, pp. 3523–3531, sep 2012.

[34] M. Elbuluk and M. Kankam, “Potential starter/generator technologies for future aerospace applications,” IEEE Aerospace and Electronic Systems Magazine, vol. 12, pp. 24–31, may 1997.

[35] T. Kostakis, P. J. Norman, and S. J. Galloway, “Assessing network architectures for the more electric engine and aircraft,” in 2014 49th International Universities Power Engineering Conference (UPEC), pp. 1–6, IEEE, sep 2014.

[36] A. Jack, B. Mecrow, and J. Haylock, “A comparative study of permanent magnet and switched reluctance motors for high performance fault tolerant applications,” Industry Applications, vol. 1, no. 4, pp. 889–895, 1996.

115 [37] B. Mecrow, a.G. Jack, J. Haylock, and J. Coles, “Fault-tolerant permanent magnet machine drives,” IEE Proceedings - Electric Power Applications, vol. 143, no. 6, p. 437, 1996.

[38] B. C. Mecrow, A. G. Jack, D. J. Atkinson, S. Green, G. J. Atkinson, A. King, and B. Green, “Design and testing of a 4 phase fault tolerant permanent magnet machine for an engine fuel pump,” in IEMDC 2003 - IEEE International Electric Machines and Drives Conference, vol. 2, pp. 1301–1307, 2003.

[39] C. R. Avery, S. G. Burrow, and P. H. Mellor, “Electrical generation and distribution for the more electric aircraft,” in 2007 42nd International Universities Power Engineering Conference, pp. 1007–1012, IEEE, sep 2007.

[40] S. MacMinn and W. Jones, “A very high speed switched-reluctance starter-generator for aircraft engine applications,” in Proceedings of the IEEE National Aerospace and Electronics Conference, pp. 1758–1764, IEEE.

[41] A. Radun, “High-power density switched reluctance motor drive for aerospace applications,” IEEE Transactions on Industry Applications, vol. 28, no. 1, pp. 113–119, 1992.

[42] M. van der Geest, H. Polinder, J. A. Ferreira, and D. Zeilstra, “Machine selection and initial design of an aerospace starter/generator,” in 2013 International Electric Machines & Drives Conference, pp. 196–203, IEEE, may 2013.

[43] I. Alan and T. Lipo, “Control of a polyphase induction generator/induction motor power conversion system completely isolated from the utility,” IEEE Transactions on Industry Applications, vol. 30, no. 3, pp. 636–647, 1994.

[44] S. Sul and T. Lipo, “Field-oriented control of an induction machine in a high frequency link power system,” IEEE Transactions on Power Electronics, vol. 5, no. 4, pp. 436–445, 1990.

[45] C. M. Taylor, “Electric Starting of Large Aircraft Engines,” in SAE Technical Paper, nov 2002.

[46] A. Eid, H. El-Kishky, M. Abdel-Salam, and T. El-Mohandes, “Modeling and character- ization of an aircraft electric power system with a fuel cell-equipped APU paralleled at main AC bus,” in 2010 IEEE International Power Modulator and High Voltage Confer- ence, pp. 229–232, IEEE, may 2010.

[47] K. Cheng, “Comparative study of AC/DC converters for More Electric Aircraft,” in Seventh International Conference on Power Electronics and Variable Speed Drives, vol. 1998, pp. 299–304, IEE, 1998.

116 [48] Xin Zhao, J. M. Guerrero, and Xiaohua Wu, “Review of aircraft electric power systems and architectures,” in 2014 IEEE International Energy Conference (ENERGYCON), pp. 949–953, IEEE, may 2014.

[49] X. Roboam, B. Sareni, and A. Andrade, “More Electricity in the Air: Toward Optimized Electrical Networks Embedded in More-Electrical Aircraft,” IEEE Industrial Electronics Magazine, vol. 6, pp. 6–17, dec 2012.

[50] I. Moir, “More-electric aircraft - system considerations,” in IEE Colloquium. Electrical Machines and Systems for the More Electric Aircraft, vol. 1999, pp. 10–10, IEE, 1999.

[51] I. Moir, “The all-electric aircraft - major challenges,” in IEE Colloquium on All Electric Aircraft, vol. 1998, pp. 2–2, IEE, 1998.

[52] M. Olaiya, “High power variable frequency generator for large civil aircraft,” in IEE Colloquium. Electrical Machines and Systems for the More Electric Aircraft, vol. 1999, pp. 3–3, IEE, 1999.

[53] B. H. Nya, J. Brombach, and D. Schulz, “Beneﬁts of higher voltage levels in aircraft electrical power systems,” in 2012 Electrical Systems for Aircraft, Railway and Ship Propulsion, pp. 1–5, IEEE, oct 2012.

[54] J. C. Shaw, S. D. A. Fletcher, P. J. Norman, and S. J. Galloway, “More electric power system concepts for an environmentally responsible aircraft (N+2),” in 2012 47th In- ternational Universities Power Engineering Conference (UPEC), pp. 1–6, IEEE, sep 2012.

[55] R. D. Telford, S. J. Galloway, and G. M. Burt, “Evaluating the reliability & availability of more-electric aircraft power systems,” in 2012 47th International Universities Power Engineering Conference (UPEC), pp. 1–6, IEEE, sep 2012.

[56] J. Brombach, T. Schroter, A. Lucken, and D. Schulz, “Optimized cabin power supply with a +/ 270 V DC grid on a modern aircraft,” in 2011 7th International Conference- Workshop Compatibility and Power Electronics (CPE), pp. 425–428, IEEE, jun 2011.

[57] G. Whyatt and L. Chick, “Electrical Generation for More-Electric Aircraft using Solid Oxide Fuel Cells,” tech. rep., Paciﬁc Northwest National Laboratory, 2012.

[58] D. van den Bossche, “The A380 Flight Control Electrohydrostatic Actuators, Achieve- ments and Lessons Learnt,” in 25th Congress of International Council of the Aeronau- tical Sciences, 2006.

[59] A. R. Behbahani and K. J. Semega, “Control Strategy for Electro-Mechanical Actuators Versus Hydraulic Actuation Systems for Aerospace Applications,” nov 2010.

117 [60] M. E. Elbuluk and M. D. Kankam, “Motor drive technologies for the Power-By-Wire (PBW) program: options, trends and tradeoﬀs. II. Power electronic converters and devices,” IEEE Aerospace and Electronic Systems Magazine, vol. 10, p. 31, dec 1995.

[61] D. Trainer, “Electric actuation - power quality management of aerospace ﬂight control systems,” in International Conference on Power Electronics Machines and Drives, vol. 2002, pp. 229–234, IEE, 2002.

[62] S. C. Jensen, G. D. Jenney, B. Raymond, and D. Dawson, “Flight Test Experience with an Electromechanical Actuator on the F-18 Systems Research Aircraft,” tech. rep., 2000.

[63] K. Ahlstrom and J. Torin, “Future architecture of ﬂight control systems,” IEEE Aerospace and Electronic Systems Magazine, vol. 17, pp. 21–27, dec 2002.

[64] M. Todeschi, “Airbus - EMAs for Flight Controls Actuation System - An Important Step Achieved in 2011,” in SAE Technical Paper, oct 2011.

[65] A. Boglietti, A. Cavagnino, A. Tenconi, S. Vaschetto, and P. di Torino, “The safety critical electric machines and drives in the more electric aircraft: A survey,” in 2009 35th Annual Conference of IEEE Industrial Electronics, pp. 2587–2594, IEEE, nov 2009.

[66] F. Ciancetta, N. Rotondale, A. del Pizzo, and M. Nardi, “Development of a measure- ment system to test the eﬃciency of electrical generators for energy recovery on aircraft actuators,” in 2013 International Conference on Clean Electrical Power (ICCEP), pp. 564–570, IEEE, jun 2013.

[67] R. Crowder, “Electrically powered actuation for civil aircraft,” in IEE Colloquium on Actuator Technology: Current Practice and New Developments, vol. 1996, pp. 5–5, IEE, 1996.

[68] U.S. Department of Defense, “MIL-STD-704 Aircraft Electric Power Characteristics,” tech. rep., 2004.

[69] A. Burdette, M. Goodnow, and C. Singer, “An Overview of MIL-STD-704, MIL-HDBK- 704, 28 Volt DC Aircraft Utilization Compliance Testing, Electrical Power Quality Causes and Impacts,” nov 2004.

[70] C. I. Hill, S. Bozhko, T. Tao Yang, P. Giangrande, and C. Gerada, “More Electric Aircraft Electro-Mechanical Actuator Regenerated Power Management,” in 2015 IEEE 24th International Symposium on Industrial Electronics (ISIE), pp. 337–342, IEEE, jun 2015.

[71] T. X. Wu, J. Zumberge, and M. Wolﬀ, “On regenerative power management in more electric aircraft (MEA) power system,” in Proceedings of the 2011 IEEE National Aerospace and Electronics Conference (NAECON), pp. 211–214, IEEE, jul 2011.

118 [72] N. Madavan, “Hybrid-Electric and Distributed Propulsion Technologies for Large Com- mercial Air Transports: A NASA Perspective,” in IEEE Energy Conversion Congress & EXPO Special Session on Future Electric aircraft, 2015.

[73] M. K. Bradley and C. K. Droney, “Subsonic Ultra Green Aircraft Research: Phase I Final Report,” tech. rep., 2011.

[74] M. K. Bradley and C. K. Droney, “Subsonic Ultra Green Aircraft Research PhaseII: N+4 Advanced Concept Development,” tech. rep., 2012.

[75] C. Pornet, C. Gologan, P. C. Vratny, A. Seitz, O. Schmitz, A. T. Isikveren, and M. Hor- nung, “Methodology for Sizing and Performance Assessment of Hybrid Energy Aircraft,” Journal of Aircraft, vol. 52, pp. 341–352, jan 2015.

[76] C. Pornet and A. Isikveren, “Conceptual design of hybrid-electric transport aircraft,” Progress in Aerospace Sciences, vol. 79, pp. 114–135, 2015.

[77] M. Chomat, ed., New Applications of Electric Drives. InTech, dec 2015.

[78] C. Luongo, P. Masson, T. Nam, D. Mavris, H. Kim, G. Brown, M. Waters, and D. Hall, “Next Generation More-Electric Aircraft: A Potential Application for HTS Supercon- ductors,” IEEE Transactions on Applied Superconductivity, vol. 19, pp. 1055–1068, jun 2009.

[79] F. G. Harmon, A. A. Frank, and J.-J. Chattot, “Conceptual Design and Simulation of a Small Hybrid-Electric Unmanned Aerial Vehicle,” Journal of Aircraft, vol. 43, pp. 1490– 1498, sep 2006.

[80] R. Bansal, “Three-Phase Self-Excited Induction Generators: An Overview,” IEEE Transactions on Energy Conversion, vol. 20, pp. 292–299, jun 2005.

[81] D. Pan, Y. Wang, and T. A. Lipo, “A series regulated open-winding PM generator based constant voltage, variable frequency AC distribution system,” in 2013 IEEE ECCE Asia Downunder, pp. 214–220, IEEE, jun 2013.

[82] D. W. Novotny and T. A. Lipo, Vector Control and Dynamics of AC Drives. Oxford University Press, 1996.

[83] R. Leidhold, G. Garcia, and M. Valla, “Induction generator controller based on the instantaneous reactive power theory,” IEEE Transactions on Energy Conversion, vol. 17, pp. 368–373, sep 2002.

[84] T. Ahmed, K. Nishida, and M. Nakaoka, “Advanced control of PWM converter with variable-speed induction generator,” IEEE Transactions on Industry Applications, vol. 42, pp. 934–945, jul 2006.

119 [85] H. Akagi, E. H. Watanabe, M. Aredes, and Institute of Electrical and Electronics Engi- neers., Instantaneous power theory and applications to power conditioning. Wiley-IEEE, 2007.

[86] G. Gallegos-Lopez, F. S. Gunawan, and J. E. Walters, “Current Control of Induction Machines in the Field-Weakened Region,” in Industry Applications Conference, 2006. 41st IAS Annual Meeting. Conference Record of the 2006 IEEE, vol. 1, pp. 104–110, 2006.

[87] V. Kaura and V. Blasko, “Operation of a phase locked loop system under distorted utility conditions,” IEEE Transactions on Industry Applications, vol. 33, no. 1, pp. 58– 63, 1997.

[88] C. Lascu, I. Boldea, and F. Blaabjerg, “A modiﬁed direct torque control for induction motor sensorless drive,” IEEE Trans. Ind. Appl., vol. 36, no. 1, pp. 122–130, 2000.

[89] H. Stemmler and P. Guggenbach, “Conﬁgurations of high-power voltage source inverter drives,” in Power Electronics and Applications, 1993., Fifth European Conference on, 1993.

[90] Y. Jia, U. R. Prasanna, and K. Rajashekara, “An open-end winding induction generation system for frequency insensitive AC loads in more electric aircraft,” in IECON 2014 - 40th Annual Conference of the IEEE Industrial Electronics Society, pp. 410–416, IEEE, oct 2014.

[91] Y. Jia and K. Rajashekara, “An induction generator based AC/DC hybrid electric power generation system for more electric aircraft,” in 2015 IEEE Industry Applications Society Annual Meeting, pp. 1–7, IEEE, oct 2015.

[92] J. Hong, H. Lee, and K. Nam, “Charging Method for the Secondary Battery in Dual- Inverter Drive Systems for Electric Vehicles,” IEEE Transactions on Power Electronics, vol. 30, pp. 909–921, feb 2015.

[93] J. Kim, J. Jung, and K. Nam, “Dual-Inverter Control Strategy for High-Speed Operation of EV Induction Motors,” IEEE Transactions on Industrial Electronics, vol. 51, pp. 312– 320, apr 2004.

[94] S. Yu and E. Ganev, “Next Generation Power and Thermal Management System,” in SAE International Journal of Aerospace, vol. 1, pp. 2008–01–2934, nov 2008.

[95] I. I. Moir and A. G. A. G. Seabridge, Aircraft systems : mechanical, electrical, and avionics subsystems integration. Wiley, 2008.

[96] Y. Jia and K. Rajashekara, “An Induction Generator based Auxiliary Power Unit for Power Generation and Management System for More Electric Aircraft,” in IEEE Energy Conversion Congress & EXPO, 2016.

120 [97] T. M. Jahns, “Improved Reliability in Solid-State AC Drives by Means of Multiple Independent Phase Drive Units,” IEEE Transactions on Industry Applications, vol. IA- 16, pp. 321–331, may 1980.

[98] T. Gopalarathnam, H. Toliyat, and J. Moreira, “Multi-phase fault-tolerant brushless DC motor drives,” in Conference Record of the 2000 IEEE Industry Applications Conference. Thirty-Fifth IAS Annual Meeting and World Conference on Industrial Applications of Electrical Energy (Cat. No.00CH37129), vol. 3, pp. 1683–1688, IEEE.

121 BIOGRAPHICAL SKETCH

Yijiang Jia received his B.S degree in Automation from Beijing University of Technology, Beijing, China P. R. in 2008 and M.S. degree in Electrical Engineering from Illinois Institute of Technology, Chicago in 2012. He worked as an Electronics Mechanical Design Engineer at GE Transportation from 2012 to 2013. He is currently pursuing his Ph.D. degree at The University of Texas at Dallas. His research interests include electrical power generation system architectures for more electric aircraft and electrical motor drives using wide band gap devices.

122 CURRICULUM VITAE

Yijiang Jia October 15, 2016

Contact Information: Department of Electrical Engineering Voice: (813) 340-0883 The University of Texas at Dallas Email: [email protected] 800 W. Campbell Rd. Richardson, TX 75080-3021, U.S.A. Educational History: B.S., Automation, Beijing University of Technology, 2008 M.S., Electrical Engineering, Illinois Institute of Technology, 2012 Ph.D., Electrical Engineering, The University of Texas at Dallas, 2016

Employment History: Teaching/Research Assistant, The University of Texas at Dallas, September 2013 – present Electronic and Mechanical Engineer, GE Transportation, August 2012 – May 2013 Research & Development Intern, Hybrid Electric Vehicle Technology, Inc., June 2012 – August 2012 Research Assistant, Illinois Institute of Technology, March 2011 – May 2012

Professional Memberships: Institute of Electrical and Electronics Engineers (IEEE), 2014–present