Neural Network Controllers in DTC of Synchronous Motor Drives

Abstract

In recent times, permanent magnet synchronous motors (PMSM) have gained numerous industrial applications, because of simple structure, high efficiency and ease of maintenance. But these motors have a nonlinear mathematical model. To resolve this problem several studies have suggested the application of vector control (VC) and direct torque control (DTC) with soft-computing (SC) techniques. This paper presents neuro direct torque control (NDTC) of PMSM. Hence this paper aims to present a technique to control speed and torque with reduced ripple compared to previous techniques. The outputs of Artificial Neural Network(ANN) controller mechanism is compared with that of classical DTC and the results demonstrate the influence of ANN is improved compared to classical DTC topology. The system is also verified and proved to be operated stably with reduced torque ripple, very low speed, sudden speed reversals, at low torque and at high torque. The proposed method validity and effectiveness has been verified by computer simulations using Matlab/Simulink®. These results are compared with the ones obtained with a classical DTC using PI speed controller.

DOI: http://dx.doi.org/10.11591/ijpeds.v3i3.4023