Non-centralized Optimization-Based Control Schemes for Large-Scale Energy Systems (Springer Theses)

✍ Scribed by W. Wicak Ananduta

• Publisher: Springer
• Year: 2021
• Tongue: English
• Leaves: 164
• Category: Library

✦ Synopsis

This book describes the development of innovative non-centralized optimization-based control schemes to solve economic dispatch problems of large-scale energy systems. Particularly, it focuses on communication and cooperation processes of local controllers, which are integral parts of such schemes. The economic dispatch problem, which is formulated as a convex optimization problem with edge‐based coupling constraints, is solved by using methodologies in distributed optimization over time-varying networks, together with distributed model predictive control, and system partitioning techniques. At first, the book describes two distributed optimization methods, which are iterative and require the local controllers to exchange information with each other at each iteration. In turn, it shows that the sequence produced by these methods converges to an optimal solution when some conditions, which include how the controllers must communicate and cooperate, are satisfied. Further, it proposesan information exchange protocol to cope with possible communication link failures. Finally, the proposed distributed optimization methods are extended to the cases with random communication networks and asynchronous updates. Overall, this book presents a set of improved predictive control and distributed optimization methods, together with a rigorous mathematical analysis of each proposed algorithms. It describes a comprehensive approach to cope with communication and cooperation issues of non-centralized control schemes and show how the improved schemes can be successfully applied to solve the economic dispatch problems of large-scale energy systems.

✦ Table of Contents

Supervisor’s Foreword
Acknowledgements
Contents
Notations
1 Introduction
1.1 Development of Energy Systems
1.2 Non-centralized Model Predictive Control
1.3 Research Questions
1.4 Thesis Outline
1.5 Other Publication
References
2 Non-centralized MPC-Based Economic Dispatch
2.1 Model of Large-Scale Energy Systems
2.2 Economic Dispatch Problem
2.2.1 Optimization Problem Formulation
2.2.2 Problem Reformulation Based on the Network of Supra-Nodes
2.3 Non-centralized MPC Scheme
2.4 Benchmark Case
2.5 Summary
References
3 Distributed Augmented Lagrangian Methods
3.1 Augmented Lagrangian Methods
3.2 Distributed Augmented Lagrangian Algorithm
3.2.1 Algorithm Design
3.2.2 Convergence Analysis
3.3 Distributed ADMM Algorithm
3.3.1 Algorithm Design
3.3.2 Convergence Analysis
3.4 Summary
References
4 Mitigating Communication Failures in Distributed MPC Schemes
4.1 Introduction
4.2 Impact of Communication Failures
4.3 Information-Exchange Protocol
4.4 Graph Selection and Reconfiguration
4.5 Case Study
4.6 Summary
References
5 Distributed Methods with Stochastic Communication
5.1 Introduction
5.2 Random Communication
5.3 Stochastic Distributed Algorithms
5.4 Convergence Analysis
5.4.1 Convergence of Stochastic DAL Algorithm
5.4.2 Convergence of Stochastic Distributed ADMM Algorithm
5.5 Numerical Simulations
5.6 Summary
References
6 Implementation of Distributed Methods for the MPC-Based Dispatch Scheme
6.1 Accelerated Versions
6.1.1 Adapting Nesterov's Accelerated Gradient Method
6.1.2 Convergence Speed Improvement
6.2 Warm Start Method
6.3 Simulation Study of DMPC Scheme
6.3.1 Simulation Setups
6.3.2 Simulation Results
6.4 Summary
References
7 A Non-centralized Control Scheme Based on Online Partitioning
7.1 Introduction
7.2 Definitions
7.3 Online Repartitioning Scheme
7.4 Coalition-Based Economic Dispatch Scheme
7.4.1 Coalition Formation
7.4.2 Coalition-Based Economic Dispatch
7.5 Suboptimality and Communication Complexity
7.6 Numerical Study
7.7 Summary
References
8 Resiliency of Non-centralized MPC Schemes Against Adversaries
8.1 Non-compliance in Non-centralized Schemes
8.2 System and Adversary Models
8.3 Robustification Against Attacks
8.3.1 Computing Probabilistic Bounds
8.3.2 Robust Reformulation
8.4 Attack Identification and Mitigation Method
8.4.1 Attack Detection
8.4.2 Identification and Mitigation Method
8.5 The Overall Scheme
8.6 Analysis of the Identification Method
8.6.1 The Case Where F=1
8.6.2 The Case Where F > 1
8.7 Case Study
8.7.1 Performance of the Overall Scheme
8.7.2 Performance of the Attack Detection Scheme
8.7.3 Discussions
8.8 Summary
References
9 Concluding Remarks
9.1 Contributions
9.2 Answering Research Questions
9.3 Directions of Future Research
Reference
Appendix A Proofs
A.1 Proof of Lemma 3.2摥映數爠eflinkle:ineqspsopt3.20
A.2 Proof of Lemma 5.2摥映數爠eflinkle:Vspstildespsadmm5.20
A.3 Proof of Theorem 5.3摥映數爠eflinkth:convspsadmmspstv5.30
A.4 Proof of Theorem 5.4摥映數爠eflinkth:ratespsadmm5.40