A fast and robust sliding mode controller for automatic voltage regulators in electrical power systems

Abstract

In this paper, a fast and robust sliding mode controller (SMC) based on a novel multi-variable sliding function is proposed for Automatic Voltage Regulators (AVRs) in electrical power systems. Lyapunov stability theorem is used to guarantee the asymptotic stability of the proposed controller. As the proposed controller is multi-variable, there may be different optimum controller parameter sets that result in performance close to each other. Hence, two optimum sets of parameters are explored by two well-known algorithms, namely, Sine Cosine Algorithm and Grey Wolf Optimization. A new objective function is considered for the optimizations, including an overshoot constraint. The performance of the proposed controller with the optimum parameter sets is evaluated for five different parameter uncertainties and load disturbance scenarios, and the resulting time domain measurements are compared with the results of recently published SMCs. A 38.36 % improvement in settling time was observed compared to the most recent study using the same realistic AVR model. In addition, the proposed controller suppressed different load disturbances, including parasitic ones, and quickly recovered the AVR output. All results revealed that the proposed controller achieved the best performance in terms of speed and robustness compared to existing controllers, despite the controller and exciter limitations.