Genetic algorithm-based optimization of PID controller parameters for DC motor speed control

This study investigated the optimization of PID controller parameters for DC motor speed control using genetic algorithms, comparing the results with the traditional Ziegler-Nichols tuning method. A comprehensive mathematical model of a DC motor was developed in MATLAB/Script, incorporating realistic operating constraints and dynamic load variations to simulate practical industrial applications. The genetic algorithm was configured with a crossover rate of 90%, mutation rate of 3%, population size of 20 individuals, and single-point crossover method, running for 100 iterations. The fitness function incorporated multiple performance criteria including total error, settling time, and overshoot. The GA-optimized PID controller with parameters kp=95.984, ki=0.250, and kd=13.005 outperformed the Ziegler-Nichols-tuned controller (kp=20, ki=0.5, kd=3) across all performance metrics. Most notably, the overshoot was reduced by approximately 59% (from 4.724% to 1.922%), while maintaining comparable rise time. The controller also demonstrated excellent robustness when subjected to variable load torques of 20 N·m and 10 N·m at different points during simulation.