A simplified adaptive neuro-fuzzy inference system control strategy for PMSM drives

Abstract

In this paper, a simplified Adaptive Neuro-Fuzzy Inference System (ANFIS)-based control strategy is presented for a vector controlled Permanent Magnet Synchronous Motor (PMSM) drive. The ANFIS uses the speed error as the only input, with three membership functions and thus only three rules. Offline training is performed for the ANFIS to learn to output the appropriate values of torque producing current isq based on the speed error. Results of MATLAB simulation show excellent speed response of the PMSM drive in both transient and steady state conditions.