Lmpm Position Controller Parameter Optimization Using Genetic Algorithm

Genetic algorithms are one of the most well-known evolutionary computing methods using heuristic searches that mimic the process of natural evolution. These methods provide very useful results to optimization and search problems. There are several factors which have particular impact on algorithm to be successful. That means correct population initialization, sufficient simulation time and optimization function – fitness function that plays very important role in description of sought extreme. We used genetic algorithm to design high performance position controller for LMPM drive. The controller should have faster dynamics, smaller position error and improved noise immunity compared to controller designed by Pole placement method. Pole placement method is a standard method of controller design which compares calculated denominator of closed-loop system to desired denominator of equal powers. Easiness of using MATLAB and Genetic algorithm toolbox is demonstrated by solving the controller design problem. Finally, results gathered from simulation in MATLAB-Simulink are presented.