A Neural Network Based Space Vector PWM for Permanent Magnet Synchronous Motor Drive

This paper presents a artificial neural network based implementation of space vector pulse width modulation (SVPWM) for voltage-source inverter(VSI) permanent magnet synchronous motor(PMSM) drive. This ANN-SVPWM completely covers the undermodulation and overmodulation modes using method of linear modulation. Different network architectures are investigated keeping the same number of neurons. A optimal neural network is obtained. The vector control system of PMSM with proposed ANN-SVPWM is simulated using MATLAB/Simulink together with the Neural Network Toolbox. The simulation results demonstrate the feasibility of the optimal network presented.