Speed Control of DC Motors Using an Artificial Neural Network (ANN)

In this paper, design and simulation of a speed controller for separately excited dc motors (SEDC) using artificial neural network (ANN) is presented. This study was conducted as a research project for undergraduate engineering education. Excellent control of speed for acceleration and deceleration, reliability, and favorable cost make the SEDC motors attractive to many applications. However, conventional control techniques are slow and generates larger overshoot which results in undesirable speed and torque ripple contributing to acoustic noise. ANN control techniques aim to minimize this ripple and noise. In this work, a novel ANN control methodology and results from a comparison study with PI control are described. This research work contributes significantly to our hands-on teaching techniques in engineering.