Neural network vector control of a permanent magnet synchronous motor drive

After analyzing the basic principle of vector control in permanent magnet synchronous motor (PMSM) drive, this paper proposes a novel artificial neural network (ANN) based vector control. Here, ANN is used in speed control and space vector pulse width modulation (SVM), since an artificial neural network based speed controller does not rely on the accurate mathematical model of system, and it has not only fast dynamic response but also high steady-state accuracy, while an artificial neural network based SVM (ANN-SVM) algorithm can be realized easily with a small amount of calculation and efficiently-reduced current harmonic. A PMSM drive simulation model with ANN based vector control is created and studied using Matlab/Simulink. The simulation results demonstrate the feasibility and validity of ANN based vector control.