Optimization tuning of PI controller of quadruple tank process

Abstract

In this paper, a swarm intelligence method, particle swarm optimization (PSO) is presented for determining the optimal proportional-integral (PI) controller parameters of a quadruple tank process (QTP). In QTP, when the multivariable zero is in the right half plane (RHP), the system becomes a nonminimum phase system which makes the tuning of controller difficult since limitation occur for linear feedback design. To overcome the limitation, the system's performance during minimum phase and non-minimum phase is investigated for a decentralized PI-tuning. To estimate the performance of the proposed PSO-PI controller, the performance index i.e. ITAE is utilized. From the results obtained, the proposed technique yields better performance in terms of settling time and maximum overshoot compared to the manual tuning method, hence an efficient technique to improving the step response of a quadruple tank process.