Adaptive RBF Neural Network Based on SMC for APF Control Strategy Study

Current harmonics are the major concern in modern equipment. In this paper, an adaptive radical basis function neural network (RBFNN) is proposed to deal with dynamic tracking error problems which are the mathematic model uncertain or complex for the three-phase active power filter (APF). APF is necessary to compensate the harmonics exits in the nonlinear load to maintain the supply current stabilization. The adaptive RBFNN systems are employed to approximate the unknown system function in the sliding mode controller. The simulation results of APF demonstrate the outstanding compensation performance and strong robustness.