Optimization-Based Tuning of PI Controller Parameters for DC Motor Speed Control

Abstract

Direct current (DC) motors are widely used in industrial applications due to their numerous advantages, such as high efficiency, cost-effectiveness, and adaptability. Therefore, accurate control of these motors is equally crucial. The most popular controller for regulating the speed of a DC motor is the conventional Proportional-Integral-Derivative (PID) controller. However, determining the parameters of a DC motor, developing a mathematical model, and subsequently identifying or experimentally selecting control parameters is a laborious and time-consuming process. In this study, the coefficients of the PI controller used for speed regulation of a DC motor were determined using the Particle Swarm Optimization (PSO) and Sine Cosine Algorithm (SCA) methods. The study was conducted experimentally for three different reference values and four distinct control methods, with the resulting data visualized using MATLAB. Step, sinus with offset, and sinus without offset signals were selected as reference values. The control methods employed included open-loop control, PI control, PSO-PI control, and SCA-PI control. When the results of open-loop control and optimization-based PI control were compared, it was observed that steady-state errors decreased by 91.25% and 90.41% for step reference with PSO and SCA, respectively; by 84.7% and 80.58% for sinus with offset reference with PSO and SCA, respectively; and by 76.72% and 74.75% for sinus without offset reference with PSO and SCA, respectively. Additionally, the motor demonstrated a more stable tracking of the reference values. When the PI control results were compared with PSO-PI and SCA-PI control, the steady-state error was found to decrease by an mean of 9.74% for the same reference values.