A Single Neuron PID Controller Based PMSM DTC Drive System Fed by Fault Tolerant 4-Switch 3-Phase Inverter

This paper presents an adaptive single neuron based PID speed controller to substitute the traditional PID controller usually used in the fault tolerant 4-switch 3-phase (4S3Ph) inverter fed permanent magnet synchronous motor (PMSM) direct torque control (DTC) drive system to improve the system performance. Based on the structure and principle analysis of the single neuron PID controller, a modified weight training algorithm is obtained by combining the supervisory delta learning rule with the non-supervisory Hebbian learning rule to achieve the controller parameters on-line self-study and optimization. The single neuron PID controller based 4S3Ph inverter fed PMSM DTC drive is modelled and studied by MATLAB/SIMULINK, simulation results show that both steady state and dynamic performance of the PMSM DTC drive system has been improved with stronger self-adaptivity and robustness comparing with the system utilizing traditional PID controller