Research On Decentralized Estimation And Reliable Predictive Control For Frequency Regulation Of Microgrid

The increasing demand for electricity and the worsening environmental issues have led to a constant increase in the demand for renewable energy in modern power systems.Microgrids can increase the penetration of renewable energy in the power system and provide flexible solutions for the supply-demand balance of the power system.However,during islanded operation of microgrids,due to the large amount of low-inertia renewable energy and the lack of inertia support from traditional generation sources,frequency deviation often occurs when new renewable energy is connected to the system.Currently,the reliable control of frequency deviation in microgrids has become a widely concerned issue.Model predictive control(MPC)methods,due to their advantages in dealing with multi-constraint optimization problems,have been widely used to solve the secondary frequency regulation problem of microgrids with variable constraints.However,it should be noted that the complexity of solving optimization problems with MPC will increase with the size of the model.As microgrids are a typical complex large-scale system,applying MPC generally leads to problems such as excessive computational complexity and long adjustment time.Recently,the emergence of decentralized modeling approaches and networked physical systems has provided a compromise solution to avoid the high computational complexity of predictive control algorithms.This paper uses a decentralized modeling approach to design an MPC controller to solve the frequency control problem of microgrids.The main contribution of this paper is an unknown input estimator is construct to estimate the state variables that are difficult to measure,and to consider;the bi-directional interaction robustness impact of estimation error and control error is construct to design an integrated optimization construction method for the estimator and controller.In response to the network attack problem faced by cloud-based optimization control of microgrids,a detection method and defense mechanism for covert attacks are designed,and a cross-layer attack detection and open-loop defense strategy that integrates local controller settings,network management,and cloud control configurations is proposed.Finally,the feasibility of the proposed solution is verified through numerical simulations.The main research work of this paper is as follows:(1)Dynamically modeling is perfomed for the microgrid system and the state-space model of decentralized subsystem is established based on the system state space and the characteristics of each distributed energy source.(2)Robust decentralized unknown input estimator are design to estimate the information on state variables that are difficult to obtain directly subsystem.(3)An integrated design approach of decentralized controllers and estimators is constructed to eliminate the effect of bidirectional interaction robustness between estimation error and control error.(4)A microgrid-cloud optimal control model is proposed and a reliable predictive control method integrating privacy protection,attack detection and open-loop control defense is proposed for stealthy attacks.