Optimal Fuzzy Controller Tuned by TV-PSO for Induction Motor Speed Control

This paper reports an automated procedure for the design of an optimal fuzzy logic controller to be used as an induction motor speed controller. The procedure consists of selection of a suitable well known fuzzy logic controller and tuning via particle swarm optimization optimal for the selected criteria. In this way the time required for tuning of the controller is significantly reduced in comparison with trial and error methods. As a benchmark a proportional-integral (PI) controller is used. The PI controller is tuned via the symmetrical optimum procedure, the standard procedure for tuning a speed controller of an induction motor. Simulation results are obtained via a mathematical model developed in Matlab/Simulink. Experimental verification is carried out with a laboratory model based on the DS1104 digital control card. To minimize iron losses and to provide better motor performance for low loads, flux is reduced from nominal and speed is kept below nominal. Results are presented in tables and graphics. The optimal fuzzy logic controller provides a slight practical advantage.