Application of Imperialist Competitive Algorithm in Online PI Controller

Abstract

In this paper, imperialist competitive algorithm is applied to the problem of finding the optimum value for an online PI controller. The objective is to design an adaptive controller for temperature control of non-isothermal reactor with respect to the set point variations. The gain scheduling method is employed to construct an online PI controller. The PI controller is designed in such a way that it minimizes the sum of square error. Fixed gain PI controllers are also implemented on process plant. The results among former method and non-dynamic approaches when compared indicated that the proposed model offers a remarkable performance.