Automatic Tuning of Decentralized Controllers by Swarm Intelligence

Abstract

This study is concerned with an application of particle swarm optimization (PSO) technique for automatic tuning of multi-loop PID controllers for multivariable systems by employing a comprehensive objective function. Controller optimization and tuning of a benchmark multivariable system is addressed to demonstrate merits of the proposed algorithm. Reported results indicate effectiveness of the introduced comprehensive objective function, which fines any violence from desired region defined for the time response of the system in different channels, simultaneously, in addition to the control effort used by the decentralized control system. It is shown that PSO is an effective algorithm in finding the optimal solution for the given problem. The proposed algorithm and PSO method have the ability of being applied to high dimensional industrial applications. The designer of the control system can enjoy optionality provided in the selection of desired region for time response of the system