Adaptive neural sliding mode control of active power filter using feedback linearization

In this paper, a radial basis function (RBF) neural network adaptive sliding mode control system based on feedback linearization approach is developed for the current compensation of three-phase active power filter(APF). RBF neural network is used to approximate the switch function of IGBT in APF combined with feedback linearization approach. The weights of RBF neural network are adjusted by means of adaptive method and the stability of the system can be guaranteed. With this method, the harmonic current of non-linear load can be eliminated and the quality of power system can be well improved. The advantages of the adaptive control, neural network control and sliding mode control are combined together to achieve the control task. Simulation results demonstrate that the control system has good control performance and can compensate harmonic current effectively.