VSG-Based Adaptive Optimal Frequency Regulation for AC Microgrids With Nonlinear Dynamics
- 20 February 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automation Science and Engineering
- Vol. 22 (15455955) , 1471-1482
Abstract
In this paper, an optimal frequency control approach is proposed for VSG-based AC Microgrids (MGs) to improve the frequency regulation performance by considering the nonlinear dynamics. First, a nonlinear dynamics system of the MG are analytically modelled by using the VSG controller. Following, the optimal controller is designed based on the Hamilton-Jacobi-Bellman (HJB) equation and is obtained by using the adaptive dynamic programming (ADP) method without linearizing. Then, a single neural networks (NNs) construction is used to approximate the optimal controller and cost function, simultaneously. Unlike the conventional VSG approach with constant inertia, the novel method can determine the optimal inertia in an online and adaptive way, as well as preserve a balance between the frequency regulation performance and the control cost. Finally, simulation results based on MATLAB/Simulink verify that the proposed control method improves the frequency nadir and the rate of change of the frequency (RoCoF) while DC-side energy is also drastically preserved by comparing the existing three studies. Note to Practitioners —a) Previous methods always assume that the generator can instantly generate and absorb infinite power, however, capacitance is finite in practice. This paper proposes an adaptive VSG-based method to provide a bounded threshold of inertia, which facilitates practical applications. b) The AC MG system is modeled as a nonlinear system to cope with the strong disturbance, which is more suitable for practical system. c) The controller is approximated by NN, which is designed by three physical variables, i.e., angular deviation, frequency deviation and rate of change of the frequency (RoCoF), which can be measured from the practical system. d) Small signal state space model is used to analyze the stability of the controller, which further proves the stability of the proposed method and enhances its application prospect in engineering practice.Keywords
Funding Information
- National Natural Science Foundation of China (62103316)
- China Postdoctoral Science Foundation (2022MD713783)
This publication has 33 references indexed in Scilit:
- Brain-Inspired Deep Meta-Reinforcement Learning for Active Coordinated Fault-Tolerant Load Frequency Control of Multi-Area GridsIEEE Transactions on Automation Science and Engineering, 2023
- Entirely Coupled Recurrent Neural Network-Based Backstepping Control for Global Stability of Power System NetworksIEEE Transactions on Automation Science and Engineering, 2023
- Distributed Observer-Based Event-Triggered Load Frequency Control of Multiarea Power Systems Under Cyber AttacksIEEE Transactions on Automation Science and Engineering, 2022
- Dual Heuristic Programming for Optimal Control of Continuous-Time Nonlinear Systems Using Single Echo State NetworkIEEE Transactions on Cybernetics, 2020
- A review of droop control techniques for microgridRenewable and Sustainable Energy Reviews, 2017
- Inertia response and frequency control techniques for renewable energy sources: A reviewRenewable and Sustainable Energy Reviews, 2017
- Self-Tuning Virtual Synchronous Machine: A Control Strategy for Energy Storage Systems to Support Dynamic Frequency ControlIEEE Transactions on Energy Conversion, 2014
- Power System Stabilization Using Virtual Synchronous Generator With Alternating Moment of InertiaIEEE Journal of Emerging and Selected Topics in Power Electronics, 2014
- Virtual synchronous generators: A survey and new perspectivesInternational Journal of Electrical Power & Energy Systems, 2014
- Online actor–critic algorithm to solve the continuous-time infinite horizon optimal control problemAutomatica, 2010