Evaluation function comparison of particle swarm optimization for buck converter

PID controller has been frequently applied to various application areas because of mathematical definition and overall performance which is greatly depended on accuracy of the control parameters. Therefore, conventional and evolutionary algorithms have introduced to find these parameters as tuning methods. In present evolutionary algorithms have frequently used in PID tuning applications. However, the tuning performance greatly depends on the evaluation function. This study focuses on design of a particle swarm optimization (PSO) based PID controller by using 8 different fitness (evolution) functions. The performance of the optimized controllers is compared with respect to following criteria: Overshoot, Undershoot, Rise Time, Settling Time, and Steady State Error. To compare the proposed optimized controller, a DC-DC buck converter is selected as a test bed. Firstly, the parameters are optioned in the simulation environment, and then the optimized controllers are compared in the hardware circuit.