Design of Optimal FOPID Controller for Unstable Magnetic Levitation System Using PSO and ACO Techniques

Abstract

The main goal of the proposed fractional order PID controller optimized by particle swarm and ant colony optimization algorithms is to control the ferromagnetic ball position to track the reference position signal by minimizing the integral time absolute error of the levitated object. In addition to this, the designed controller improves the response characteristics of the system like overshoot, rise time, settling time, peak response, and steady state error. To stabilize an unstable maglev system, the designed compensator parameters are determined by the Routh Hurwitz stability criterion. To improve the closed loop performance of the designed FOPID controller, the system controlled by the proposed PSO-FOPID controller results a minimized percentage overshoot, settling time, peak response, rise time and steady state error from (5.851% to 2.577%), (18.898ms to 4.034ms), (1.063cm to 1.023cm), (5.925sec to 816.434micro sec) and (0.0001 to 0) as compared to FOPID controller. Finally, for improved position control and stabilization of magnetic levitation system, ACO optimized FOPID controller is designed. Furthermore, a comparison between MATLAB\Simulink simulation results of the magnetic levitation system controlled by FOPID, PSO optimized FOPID and ACO optimized FOPID controllers are made to check the performance of the designed controllers for magnetic levitation system. The FOPID controller parameters are more optimized by ant colony optimization algorithm; magnetic levitation system has good performance when it is controlled by ACO optimized FOPID controller. An ant colony optimization based fractional order PID controller gives improved overshoot, peak response, and settling time from (2.577% to 0.223%), (1.023cm to 1.003cm), (4.034msec to 1.472msec) than PSO optimized fractional PID controller.