Evolutionary Design of Fuzzy Controllers
Abstract
Introduction: PID controllers are well known and most widely used in the industries due to their simple structure and easy implementation. Nonetheless, linear PID controllers sometimes do not provide good quality of control over non-linear objects. Purpose: The goal of this study is building evolutionary-adjusted fuzzy PID controllers to improve the control performance of the conventional PID controllers. Results: A fuzzy PID controller has been studied with three separate rule bases. A two-stage scheme is proposed to adjust the controller for a nonlinear dynamic plant. At the first stage, a genetic algorithm is used to adjust a linear PID controller. The obtained PID coefficients are used as fuzzy PID output scaling factors. At the second stage, the genetic algorithm is used to form a nonlinear mapping function for every channel, implemented on the base of an artificial neural network. The proposed control algorithm has been validated with MatLab. The obtained results demonstrate that the proposed controller provides excellent dynamic and steady-state characteristics as compared to the traditional controllers. Practical relevance: Fuzzy PID controllers are suitable for the control over nonlinear plants in industrial applications, as demonstrated by the examples in this paper.Published
2015-12-01
How to Cite
Burakov, M., Konovalov, A., & Yakovets, O. (2015). Evolutionary Design of Fuzzy Controllers. Information and Control Systems, (6), 28-33. https://doi.org/10.15217/issn1684-8853.2015.6.28
Issue
Section
Information and control systems