Feedforward Neural Network-DTC of Multi-phase Permanent Magnet Synchronous Motor Using Five-Phase Neural Space Vector Pulse Width Modulation Strategy

Abstract

In this work, the direct torque control (DTC) is applied to the five-phase permanent magnet synchronous motor (FP-PMSM). The DTC method based on classical space vector pulse width modulation (SVPWM) is a common solution used to overcome traditional problems; such as stator flux ripple, electromagnetic torque ripple and gives more total harmonic distortion (THD) of the stator current. The actual paper is based on improving the performance of DTC-SVPWM by using the feedforward neural networks (FNNs) instead of the proportional-integral (PI) regulators and hysteresis comparators (HCs) of the conventional SVPWM strategy. This algorithm can solve the traditional PI regulators and HCs problems which are represented in responses dynamic and reduce the torque ripple, flux ripple, and the THD of stator current of FP-PMSM drives. The proposed strategy was tested in different tests with simulation using Matlab software.