A Modified GA based PI controller for DC Motor Performance

Despite there have been numerous attempts to examine various concepts of genetic algorithm to improve its performance, but there is very little attention to analyzing objectively the influence of modifying initialization constraints on GA based controller problems. The aim is to improve GA performance to become a superior algorithm for optimizing PI controller. For this reason, it was proposed a novel method by Modified Initialization Fitness Function MIFF tackled with GA, by using optimization-based tune OBT to improve initialization constraints. The main contribution of this research is that we have designed the PI controller on GA based MIFF (GA_MIFF) to improve the speed performance of the DC motor controller and to overcome the drawbacks in classical GA. The study is based on a comparison of three techniques proposed controller, classical GA and CHR-SIOS, applied to three different models of DC motors. Finally, to perform a reliable comparison of these results, the bench comparison for validation was done with classical GA in term of improvement reduction step response ratio(IRSRR).The competition performance between them was estimated in term of dead time td, rise time tr, settling time ts, where the overshoot was eliminated for both classical GA and proposed (GA_MIFF). The results show that the proposed (GA_MIFF) overcomes classical GA for all controlled models understudy.