Optimization-Driven PID Tuning for PMDC Motor: A Simulink Comparative Study of PSO, FPA, and DE

Permanent Magnet DC (PMDC) motors are commonly used in many industrial setups and research projects because they are easy to deal with, dependable in performance, and allow simple speed control. Despite their popularity, tuning a PID controller by hand still causes uneven results, especially when the motor works under different or changing conditions. In this work, a simulation model was built in MATLAB/Simulink to look for better PID gains using three well-known metaheuristic methods: Particle Swarm Optimization (PSO), Flower Pollination Algorithm (FPA), and Differential Evolution (DE). The main goal is to compare how each technique searches for the best controller settings and how much it improves the motor’s response. The mathematical model of the motor was identified from experimental PWM–RPM data and implemented in Simulink to simulate the system behavior under closed-loop control. The optimization process employed the Integral of Time-weighted Absolute Error (ITAE) as the cost function to ensure fast response, minimal overshoot, and small steady-state error. Simulation results demonstrated that all three algorithms improved the control performance compared to conventional PID tuning methods. Among them, the DE-based controller achieved the most balanced performance, reaching a steady-state error of 0.0029 with zero overshoot and smooth transient behavior. The PSO-based controller provided the fastest rise time, while FPA offered the most stable response with no oscillations. Overall, the proposed Simulink framework provides an efficient and low-cost environment for evaluating intelligent optimization techniques before their implementation on real hardware.