Robust fuzzy model predictive control for voltage regulation in islanded microgrids

Abstract

An effective robust fuzzy model predictive control (RFMPC) method for secondary voltage control in islanded microgrids (μGs) is presented here. In contrast to the existing techniques, which require a detailed model of μG and an ideal communication network between μG central controller and primary local controllers, RFMPC is synthesized for a non-linear model of the μG with various time delays, uncertainties, and bounded disturbances. The famous Takagi-Sugeno fuzzy approach is adopted to approximate the inherently non-linear model of μG by locally linear dynamics. The Lyapunov–Razumikhin functional method is exploited to deal with time delays. In this regard, sufficient conditions are provided in the form of linear matrix inequalities (LMIs). Then, a sequence of control laws corresponding to a set of terminal constraints is computed offline. Doing so, the online stage is reduced to solving a convex problem with LMI constraints considering the sequence of constraint sets obtained in the offline stage, thereby reducing the computational burden significantly. Robust positive invariance and input-to-state stability property concerning communication network deficiency are then speculated. The effectiveness of the proposed RFMPC is verified via a comprehensive suite of simulations in the MATLAB/SimPowerSystems environment.