Adaptive neural network based controller for direct torque control of PMSM with minimum torque ripples

Abstract

An Artificial Neural Network (ANN) based controller for Permanent Magnet Synchronous Motor (PMSM) under Direct Torque Control (DTC) algorithm is proposed to minimize the torque ripples associated with Hysteresis Direct Torque Control (HDTC). In this system, the stator flux position, stator flux error and developed torque error are used to select two active vectors while at the same time, the normalized absolute value of these ones are used in artificial neural network algorithm block to adapt the switching of the inverter in order to control the applied average voltage level in such a way to minimize the torque ripples. Thus, it includes the capability of ANN to learn from processes and the fast response feature of the DTC. The simulated results show considerable torque ripple and current ripple reduction as well as ElectroMagnetic Interference (EMI) noise level reduction. The simulated results are supported with some experimental results.