Design of fractional order PID controller for DC motor using evolutionary optimization techniques

摘要

In this paper, a comparison of various Evolutionary Optimization (EO) techniques has been performed for controller tuning (PID and FOPID) for DC motor speed control applications. Three techniques have been considered, which are: genetic algorithm (GA), particle swarm optimization (PSO), and differential evolution (DE). A fixed structure PID and FOPID controller has been considered. Usually in literature, comparison has been carried out based on time domain performance indices; whereas, in this work, mixed sensitivity H_∞ method is considered as the fitness function for the comparison of various EO techniques in order to assess the robustness of the designed system. It has been shown that GA optimized FOPID controller has better robustness with respect to model uncertainties.