Speed Control of DC Motor by Using Neural Network Parameter Tuner for PI-controller

Abstract

This paper proposes the improvement of the speed and torque dynamic responses of DC motor by using neural network parameter tuner (NNPT) for PI-controller. The performance of traditional PI-controllers has been insufficient in various speed controls of DC motors due to nonlinear structure, changing parameters, disturbance and various load conditions. This task can be solved by building an adaptive control system for DC motors based on the operational setting of the parameters of linear regulators. In this paper, a neural network parameters tuner for PI controller was developed, which allows adjusting the values of Ki and Kp in the conditions of changing the various speed and the parameters of the DC motor. The article presents the functional diagram of the neural network tuner, the structure of the neural network and mathematical model of DC motor. The results show that the neural network parameter tuner for PI-controller gives the superiority performance.