Design and Analysis of Distributed Energy Management Systems (eBook)

Preface 6
Contents 10
Part I Design and Analysis of Energy Management Systems Considering Consumer Demand and Use of Electric Vehicles 12
1 Activity-Based Modeling for Integration of Energy Systemsfor House and Electric Vehicle 13
1.1 Introduction 13
1.2 Energy Demand Modeling for the Residential Sector 14
1.2.1 Occupancy-Based Approach 14
1.2.2 Activity-Based Approach 15
1.2.3 Time-Based Household Energy Demand Model 16
1.2.4 Important Factors in Energy Demand Modeling 17
1.3 Energy Demand Modeling for Transportation 17
1.3.1 Trip-Based Modeling 18
1.3.2 Activity-Based Modeling 18
1.4 Case Study 20
1.4.1 In-Home Activity Model 22
1.4.2 Out-of-Home Activity Model 24
1.4.3 Results 27
1.4.3.1 Activity 27
1.4.3.2 Electricity Demand 28
1.4.3.3 EV Integration Potential 29
1.4.3.4 Influence of Geographical Location 31
1.5 Conclusion 32
References 32
2 Probabilistic Model and Prediction of Vehicle Daily Use 36
2.1 How to Use a Vehicle at Home 36
2.2 Related Works 38
2.3 Statistical Data of Daily Car Use 39
2.4 Preparation for Problem Formulation 41
2.4.1 Description of Time 41
2.4.2 Expression of Future Profile of Vehicle Use 42
2.5 Problem Formulation of Predicting a PDTT 42
2.6 Markov Model Representing Production Process of a PDTT 44
2.6.1 A PDTT and a Markov Model 44
2.6.2 Calculation of Probabilities 45
2.7 Solving Method for Predicting a PDTT by Dynamic Programming 47
2.8 Simulation Result of the PDTT Estimation 50
2.9 Conclusion 53
References 53
3 Design of a Home Energy Management System Integratedwith an Electric Vehicle (V2H+HPWH EMS) 55
3.1 Introduction 55
3.2 HPWH Model 57
3.2.1 Overview of the HPWH Model 57
3.2.2 Nomenclature 57
3.2.3 Mathematical Model of the HPWH 58
3.2.4 Piecewise Linearization of the HPWH Model 59
3.3 Model Predictive HEMS Using an In-Vehicle Storage Battery and a HPWH 60
3.3.1 Formulation of the Optimization Problem 61
3.3.2 Cost Function 64
3.3.3 Constraints on Household Electric Power Consumption 65
3.3.4 Constraints on In-Vehicle Storage Batteries 65
3.3.5 Constraints on the HPWH 65
3.4 Numerical Simulations 67
3.4.1 Simulation Settings 67
3.4.2 Simulation Results 68
3.4.3 Discussion 71
3.4.3.1 Computation Time 71
3.4.3.2 Surplus Electric Power 71
3.4.3.3 Electricity Charges 72
3.5 Conclusion 72
References 73
4 Range Extension Autonomous Driving for Electric Vehicles Based on Optimization of Velocity Profile Considering Traffic Signal Information 75
4.1 Introduction 75
4.2 Experimental Vehicle and Its Mathematical Model 76
4.2.1 Experimental Vehicle 76
4.2.2 Vehicle Model 77
4.2.3 Power Flow Model Harada2 79
4.3 Optimization of Velocity Profile Considering Traffic Signal Information 81
4.3.1 Signal Information Model 81
4.3.2 Evaluation Function and Constraint Condition 81
4.3.3 Comparison Conditions 82
4.3.3.1 Conventional Profile 1: Constant Acceleration and Deceleration with Signal Information 82
4.3.3.2 Conventional Profile 2: Optimized Velocity Profile Without Signal Information 83
4.3.3.3 Proposed Profile: Optimized Velocity Profile with Signal Information 83
4.4 Simulation Results 83
4.5 Experimental Results 87
4.5.1 Control System 87
4.5.2 Experimental Environment 87
4.5.3 Experiment Results of RC-S 89
4.5.4 Experimental Results of the Driving Test 90
4.6 Conclusion 91
References 91
Part II Synthesis of Distributed Energy Management Systems Based on Aggregation of Local EMSs and Vehicles 93
5 Real-Time Pricing and Decentralized Optimization Strategyfor Power Flow Balancing in EV/PHV Storage Management 94
5.1 Introduction 94
5.2 Real-Time Pricing and Decentralized Optimization Leading to Optimal Operation 95
5.2.1 Dynamics of Agent 95
5.2.2 Decentralized Determination of Optimal Set-Point 96
5.2.3 Real-Time Pricing Strategy 98
5.2.3.1 Pricing Strategy in Steady-State 98
5.2.3.2 Gradient Based Real-Time Pricing Strategy 99
5.2.4 Stability Analysis 100
5.2.4.1 Local Behavior of Decentralized Decision-Making 101
5.2.4.2 Uniqueness of the Equilibrium Point in Local Dynamics 102
5.2.4.3 Local Stability of the Closed-Loop System 103
5.3 EV Storage Management for Power Flow Balancing 104
5.4 Numerical Case Studies 107
5.4.1 Numerical Case Study 1: Charging and Discharging 108
5.4.2 Numerical Case Study 2: Plug-and-Play Type Operation 108
5.4.3 Numerical Case Study 3: Community Consists of 50 Vehicles 111
5.5 Conclusions 111
References 113
6 A Scalable Control Approach for Providing Regulation Services with Grid-Integrated Electric Vehicles 114
6.1 Introduction 114
6.2 Related Work 115
6.2.1 Regulation Services 115
6.2.2 Heuristic Scheduling for DR 116
6.3 Problem Formulation 116
6.3.1 Decision Phase I 117
6.3.2 Decision Phase II 118
6.3.3 Challenges in a Large-Scale Setting 120
6.4 Bin-Based GIV Control Approach 121
6.4.1 EV Charging Behavior 122
6.4.2 Binning Mechanism 124
6.4.3 Scheduling Mechanism 128
6.5 Evaluation 129
6.5.1 Alternative GIV Approaches 129
6.5.2 Simulation Results: Scheduling Quality 130
6.5.3 Simulation Results: Scheduling Scalability 131
6.6 Conclusions and Future Work 133
References 134
7 A Continuum Approach to Assessing the Impact of Spatiotemporal EV Charging to Distribution Grids 136
7.1 Introduction 136
7.2 ODE Modeling of Distribution Voltage Profile 138
7.3 Numerical Demonstration 140
7.3.1 Setting of Feeder and Charging Stations 140
7.3.2 Construction of Power Density Function 141
7.3.3 Results 142
7.4 Discussions 144
7.4.1 Comparison with Power-Flow Equation 144
7.4.2 An Analytical Treatment 145
7.5 Conclusion 146
References 147
Part III Toward Dependable Distributed Energy Management System Using ICT 149
8 Cyber Security for Voltage Control of Distribution Systems Under Data Falsification Attacks 150
8.1 Introduction 150
8.2 Voltage Regulation Based on Centralized Control 152
8.2.1 Centralized Voltage Control 152
8.2.2 Simulation Settings for Voltage Regulation 154
8.3 Cyberattacks and Security Measures in Voltage Regulation 156
8.3.1 False Data Injection (FDI) Attacks 157
8.3.2 Detection Algorithm for FDI Attacks 158
8.3.3 Stealthy Cyberattack Strategies 158
8.3.4 Further Measures for Resilient Control 160
8.4 Simulation Settings for Cyber Security 162
8.4.1 Attacker Strategy 162
8.4.2 Resilient Control 163
8.5 Verification Via Simulations 164
8.6 Discussions and Further Studies 167
8.7 Conclusion 167
References 169
9 Machine Learning Based Intrusion Detection in Control System Communication 171
9.1 Current Status for Intrusion Detection in Control System 171
9.2 Various Approaches 172
9.3 Intrusion Detection for Control System Communication Without Sequence Patterns 173
9.3.1 The Feature Representation for Intrusion Detection in Control System Communication 174
9.3.2 Binary Classification Methods 175
9.3.2.1 C4.5 175
9.3.2.2 Support Vector Machine (SVM) 176
9.3.3 Anomaly Detection Methods 178
9.3.3.1 Local Outlier Factor (LOF) 178
9.3.3.2 One-Class Support Vector Machine (OCSVM) 180
9.3.4 Support Vector Data Description (SVDD) 182
9.4 Experiments for Control System Communication Without Control Sequence Patterns 184
9.4.1 Experimental Setup 184
9.4.1.1 Water Storage Tank Control System Communication Data 184
9.4.1.2 Gas Pipeline Control System Communication Data 185
9.4.1.3 Cyberattacks in Experiments 185
9.4.2 Control System Communication Data 186
9.4.2.1 Difference Between Measured Values d 186
9.4.2.2 Alarm e 186
9.4.2.3 Data Normalization 187
9.4.2.4 ID Frequency 187
9.4.2.5 Training Data and Test Data 187
9.4.3 Evaluation Criteria 188
9.4.4 Experimental Results 188
9.5 Intrusion Detection Considering Sequences 192
9.5.1 Control Sequences for Intrusion Detection 192
9.5.2 Hidden Markov Model (HMM) 193
9.5.3 Conditional Random Field (CRF) 194
9.6 Experiments for Control System Communication with Control Sequence Patterns 196
9.6.1 Experimental Setup 197
9.6.1.1 Benchmark Data (DARPA Data) 197
9.6.1.2 Control System Communication Data by Simulator 198
9.6.2 Evaluation Criteria 200
9.6.3 Experimental Results 201
9.6.3.1 Experimental Results for Benchmark Data (DARPA Data) 201
9.6.3.2 Experimental Results for Simulated Control System Communication Data 203
9.7 Conclusion 204
References 205
Index 207