DRELIABLE MODELLING, SIMULATION, IDENTIFICATION, CONTROL AND ESTIMATION FOR DYNAMIC SYSTEM WITH UNCERTAINTY

ABSTRACT This study focuses on the reliable modelling, simulation, identification, control, and estimation of a DC motor operating under conditions of uncertainty. DC motors are widely used in various industrial applications where they are often subjected to parameter variations, external disturbances, and measurement noise. To address these challenges, this research developed a comprehensive mathematical model of the DC motor, incorporating these uncertainties, and implemented it in MATLAB/SIMULINK. The study employed Particle Swarm Optimization (PSO) to tune a Proportional-Integral (PI) controller, aiming to optimize the motor's performance under unbalanced voltage and frequency conditions. System identification techniques were used to refine the model based on simulated data, enhancing its accuracy. Robust control strategies, including PID, adaptive, and H-infinity controls, were designed and tested. Additionally, state estimation algorithms such as the Kalman Filter, Extended Kalman Filter, and Particle Filter were developed to ensure accurate state estimation despite uncertainties. Simulation results demonstrated that the PSO-tuned PI controller significantly improved the dynamic performance of the DC motor. The controller effectively minimized overshoot, reduced rise and peak times, and eliminated steady-state error, ensuring robust and reliable motor operation. The study concluded that the PSO-PI controller is superior in maintaining optimal performance under uncertain conditions, providing a viable solution for enhancing the reliability and efficiency of DC motors in industrial applications.