A Hybrid Bio-Inspired Metaheuristic Applied to PID Controller Parameter Tuning

Abstract

In this paper, a hybrid population-based, bio-inspired metaheuristic is applied to solve the PID controller parameter tuning problem. The applied algorithm, denoted by CSASOS, is a multi-population metaheuristic that combines good characteristics of the Crow Search Algorithm (CSA) and the Symbiotic Organisms Search (SOS). In CSASOS, one subpopulation follows the search rules of the CSA algorithm and is responsible for finding promising regions in the search space. The second subpopulation follows the search rules of the SOS algorithm and is responsible for performing a refined search in the promising regions to find the best solution for the optimization problem under consideration. Numerical experiments are carried out in order to compare the performance of CSASOS with the performance of the stand-alone versions of both CSA and SOS in the solution of a PDI controller parameter tuning problem, which is a common problem that appears in many industrial applications. The results show that CSASOS presented a better performance in comparison with the competing algorithms.