A Backstepping Neural Global Sliding Mode Control Using Fuzzy Approximator for Three-Phase Active Power Filter

In this paper, a backstepping global sliding mode fuzzy control based on neural network together with a proportional integral derivative (PID) sliding mode manifold is proposed for a three-phase active power filter (APF). This system consists of a backstepping PID global sliding mode controller, a fuzzy uncertainty approximator and a neural estimator. The design process of Lyapunov function and the controller can be systematic and structured through the reverse design by backstepping control. A global sliding mode controller is introduced to obtain overall robustness, speeding up the system response. On the issues of the conventional proportional derivative sliding mode surface, the integral item is added to suppress the steady state error and enhance the robustness. Besides, it is beneficial to estimate the unknown dynamic characteristics of the APF system via RBF neural network. Furthermore, the fuzzy controller utilized to approach to the switching function can reduce the influence of chattering which may lead to insulated gate bipolar transistor malfunction of the actual active power filter, achieving a better property. Finally, simulation studies with MATLAB/SimPower systems toolbox demonstrate that the designed backstepping neural PID global sliding mode fuzzy controller can obtain expected properties in three different conditions, and some comparisons are made to verify the superior performance of the raised control method.