Double Demand Responsive Electric Bus Operation to Passengers and Power System Zongxiang Wu A thesis submitted as fulfilment of the requirements for the degree of Doctor of Philosophy of Imperial College London Urban System Lab Department of Civil and Environmental Engineering Imperial College London October 2018 Declaration The research that is presented in this thesis is my own words, except where the work of others is referenced. The copyright of this thesis rests with the author. Unless otherwise indicated, its contents are licensed under a Creative Commons Attribution-Non Commercial 4.0 International Licence (CC BY-NC). Under this licence, you may copy and redistribute the material in any medium or format. You may also create and distribute modified versions of the work. This is on the condition that: you credit the author and do not use it, or any derivative works, for a commercial purpose. When reusing or sharing this work, ensure you make the licence terms clear to others by naming the licence and linking to the licence text. Where a work has been adapted, you should indicate that the work has been changed and describe those changes. Please seek permission from the copyright holder for uses of this work that are not included in this licence or permitted under UK Copyright Law. i Acknowledgements First and foremost I would like to express my sincere gratitude to my supervisors Prof John Polak and Prof Goran Strbac for their patience, motivation, and immense knowledge. Without their invaluable supports, I would not have advanced any parts of this research in transport and power system. Besides my supervisors, my sincere thank goes to Dr. Fangce Guo, Dr. Jianlin Luan, Dr Yujian Ye, Dr. Fei Teng, Dr. Dimitrios Papadaskalopoulos, Dr. Aruna Sivakumar and Dr. Konstantinos Gkiotsalitis for their insightful comments and guidence in their highly experienced field, but also for the hard question which incented me to widen my research from various perspectives. In the happy moments and the difficult ones that punctuate a four year commitment, I had the privilege to be immersed in the affection of friends and family who may not have been all physically close, but they have been all felt so. And particular gratitude to me fiancee Wenhan, wholeheartedly for your patience and accompany. ii Abstract Recently the potential for the electrification of urban bus fleets has attracted considerable policy, industrial and academic attention. Although there are significant potential benefits in terms of reduced local emissions and operating costs, electric bus operations raise a number of new challenges – how electric bus fleets and electric network operators can optimally benefit from the synergetic interaction, both as consumers of electricity and providers of grid balancing services. Researches in the public transport field exhibit simplistic learning from isolated operations in the proof-of-concept projects without considering the dynamic in electricity supply, whilst the lack of understanding of the interdependence between electric mobilities and charging flexibility has limited power system researchers applicability of load management of electric bus fleets. Against this backdrop, this thesis proposed, analysed and tested an analytic framework consist of two approaches that accommodates the interaction between the grid-integrated e-bus oper- ation and demand side management in power system. At First, the classic frequency setting models are extended to incorporate the charging dimensions – charging duration and resid- ual energy requirement - and embedded into a Stackelberg game model where the lower-level problem corresponds to the electricity pool market which generates distribution level location- specific prices. Secondly, in order to couple more complicated bus service patterns including short-turning and interlining options, a novel adaptive service scheduling approach is developed to embrace higher level of flexibility responding to the variation of passengers demand and elec- tricity tariff. Extensive case studies using real-world transit data and IEEE test system, with different scenarios of operating conditions, are used to validate the theoretical properties of the proposed mechanism. This framework provides tools enabling the electric bus fleets operator to achieve cost-effectiveness by understanding the implication of dynamics in electricity supply and electric network operators to deliver efficient load management by utilising the flexibility of electric buses loads. iii Contents Declaration i Acknowledgements ii Abstract iii Abbreviations xv 1 Introduction 1 1.1 Background ...................................... 1 1.2 Researchgapsandmotivation . ..... 3 1.3 Thesisscope..................................... 5 1.4 Thesisoutline ................................... .. 7 2 Literature Review 9 2.1 Introduction.................................... .. 9 2.2 Researchgapsinelectricbustrials . ....... 11 2.3 Load profile by operations and charging techniques . ........... 18 iv CONTENTS v 2.4 Grid-integrated application of electric bus . ............ 24 2.5 Market design for grid-integrated electric buses . ............. 30 2.6 Summary ....................................... 33 3 Grid-Integrated Electric Bus Operation in Bi-level model 36 3.1 Introduction.................................... .. 36 3.2 Assumptions on examined e-bus operational model and electricity market model 38 3.3 Literaturereviewofbusplanningproblem . ......... 42 3.4 Formulation of bus operation as frequency setting problem ............ 43 3.4.1 Formulationofchargingloadandcost . ..... 45 3.4.2 Formulation of the opportunity cost of flash charge . ......... 48 3.4.3 Formulation of the elastic passengers loads to servicequality . 49 3.4.4 Formulation of total bus operational cost . ........ 50 3.5 Formulation of lagrange-relaxation based market clearingmodel . 51 3.5.1 Losslesselectricitypoolmarket . ...... 51 3.5.2 Approximation of power loss and Loss Factor (LF) . ..... 53 3.5.3 Derivative of distribution locational marginal price (DLMP) for e-bus . 54 3.6 Bi-levelproblemandMPCCreformation . ....... 55 3.7 Implementationprocess. ..... 58 4 Case Study I: Impacts Of Dynamic Tariff On E-Bus Operation 61 4.1 Introduction.................................... .. 61 CONTENTS vi 4.2 Datapre-processing.............................. .... 62 4.2.1 General description of data and the objective of pre-processing . 62 4.2.2 Implementation of data filtering, cleaning and clustering ......... 65 4.2.3 Targetdata: Linelevelpassengerdemand . ...... 67 4.3 Descriptionofcase ............................... ... 70 4.3.1 Distributionsystemtopology . .... 70 4.3.2 Inflexibledemand............................... 73 4.3.3 Costofgenerationsystem . .. 73 4.3.4 Electricbus .................................. 74 4.3.5 Implementation ................................ 76 4.4 Validationoftheproposedframework . ........ 76 4.5 Implicationofpowernetworkcongestion . ......... 80 4.6 Sensitivity of electric mobility to cost parameters . .............. 86 4.7 Concludingremarks............................... ... 92 5 Electric Bus Multi-Service Modelling Via Adaptive Scheduling 95 5.1 Introduction.................................... .. 95 5.2 Review of approach for multiple service modelling . ............ 97 5.3 General overview of the implementation process . ........... 98 5.4 Formulation of adaptive frequency setting . ..........104 5.4.1 Mathematics formulation and ad-hoc decoding algorithm .........104 CONTENTS vii 5.4.2 Numericalexample ..............................108 5.5 Elastic passenger demand to service quality . ..........110 5.6 Charging scheme by agent-based micro-simulation (AMS) . ............111 5.7 Usagecostofbattery............................... 114 5.8 Final problem formulation: net cost minimisation . ...........115 5.9 Solution methods and conversion of constraints . ...........117 5.9.1 Untraceablefeatureofproblem . 117 5.9.2 Approximating the constrained frequency setting problem using exterior pointpenalties ................................118 5.10 StandardGenericAlgorithm(GA) . 119 5.10.1 Encoding ...................................120 5.10.2 Evaluating the fitness of individuals and selecting individuals for repro- duction ....................................121 5.10.3 Crossoverandmutation . 122 6 Case Study II: Plural Bus Service By Adaptive Scheduling And Dynamic Tariff 124 6.1 Introduction.................................... 124 6.2 Datapre-processing.............................. 125 6.2.1 Characterisethetriplength . 125 6.2.2 Controlpointsandbusloads. 130 6.3 Casestudydescription . 132 6.4 Impactsofbusservicequality . ......135 6.4.1 First look at the service patterns by adaptive scheduling .........135 6.4.2 Averagepassengerwaittimebystops . 138 6.4.3 Vehicleutilisation. 141 6.5 Impactsonchargingdemandandloads . ......142 6.6 Dependenceofe-Busoperationonbattery . ........144 6.7 Concludingremarks............................... 146 7 Conclusions And Future Research 148 7.1 Summary .......................................148 7.2 Originalcontribution . 150 7.3 Limitationandfuturework . 152 Bibliography 153 Appendices 175 A Data Brief Of Case Study 176 B Power Line Data In RBTS 179 C Load Data in RBTS 182 D Electricity Wholesale Data In Day-ahead Market 184 E Spatial And Temporal Characteristics Of Dis-aggregated Passenger Demand185 viii List of Tables 2.1 Summaries of main electric buses manufacturers . ........... 13 2.2 Summaries of electric buses trials in the world . ........... 15 4.1 General