Adaptive Fractional Sliding Mode Control of Active Power Filter Based on Dual RBF Neural Networks

Abstract

This paper presents an adaptive fractional sliding mode control scheme based on dual radial basis function (RBF) neural networks (NNs) to enhance the performance of a three-phase shunt active power filter (APF), where a conventional integer-order sliding surface is changed into a fractional-order one to speed up the system response and optimize the control performance. Furthermore, the control scheme adopts a class of dual RBF NNs, in which the network weights can be updated online to approximate the nonlinear system functions and the upper bound of estimated disturbances, respectively, improving the system stability and robustness. Meanwhile, the adaptive control laws obtained by Lyapunov analysis can guarantee the system a stable operation. Finally, by comparing with the integer-order control strategy, the simulation results verify that this proposed controller has a better performance in the suppression of the harmonic, elimination of system uncertainties, and the reduction of current tracking error.