Fuzzy Multiple Hidden Layer Neural Sliding Mode Control of Active Power Filter With Multiple Feedback Loop

A fuzzy multiple hidden layer neural sliding mode control with feedback loop (FMHLNSMCMFL) is proposed for a single-phase active power filter (APF), where controller designed to make the current tracking error converge zero and new network loops introduced approximate unknown dynamics. At same time, can eliminate chattering, improve accuracy reduce distortion rate of APF. Moreover, double recurrent weighted sum has strong global learning ability. The adaptive parameters obtained by Lyapunov function ensure asymptotic stability system. Simulation hardware experiments verify FMHLNSMCMFL scheme viable solution