Co-Design of Improved Fuzzy Switching-Type State-FDI Estimation and Resilient Control for DC Microgrids Under Malicious Attacks

DC microgrids (DC-MGs) consist of multiple units interconnected in a network. This makes the system vulnerable to malicious cyber attacks. Malicious attacks here refer to hybrid attacks of false data injection (FDI) attacks and denial of service (DoS) attacks. The main focus of this paper is to study the state-FDI estimation and resilient control problems of nonlinear DC-MGs under malicious attacks. Since the fuzzy controller/observer based on the parallel distributed compensation method cannot obtain more relaxed satisfactory results. Methods presented in the literature will be reviewed and a method based on multi-mode fuzzy switching mechanism (MMFSM) will be developed to relax the stability analysis of these DC-MGs. Furthermore, a co-design method based on improved fuzzy switching technology is proposed. This method adopts a novel MMFSM, which makes the constraints more relaxed, the state estimation more accurate, and the conservatism of the co-design method is reduced. Finally, the effectiveness of the proposed improved co-design method is verified by comparison and analysis of simulation examples. Note to Practitioners -The motivation of this paper comes from the problem that DC-MG systems, which can realize large-scale application of new energy technology, is vulnerable to malicious cyber attacks in practical tasks. Malicious attacks (such as FDI attacks and DoS attacks) will bring security risks such as data tampering and function destruction to the system. To this end, this paper develops a MMFSM-based co-design method for DC-MGs under malicious attacks. The improved co-design method can deal with sector nonlinearity, state imperfectly measurable, FDI attacks and DoS attacks simultaneously. For this method, the main difficulty in this paper is how to reduce its conservatism. To solve this problem, we adopt a novel MMFSM. By introducing time-varying equilibrium matrix synchronously with the system dynamics of different modes, the improved co-design method can switch accordingly, thereby making the constraints more relaxed and the state estimation more accurate.