An Artificial Neural Network Based SVPWM Controller for PMSM Drive

In order to research the influence of hidden layer neurons of artificial neural network (ANN) and switching frequency of power switches on the performance of permanent magnet synchronous motor(PMSM), a algorithm of space vector pulse width modulation based on artificial neural network (ANN-SVPWM) is proposed. The simulation and experiment of closed-loop PMSM control system are done. The results show that the PMSM generates less current harmonic distortion and pulsating torque by choosing the optimum hidden neurons of ANN and switching frequency of power switches, and the PMSM controlled by ANN-SVPWM works well.