Title: A data-driven PID control system using particle swarm optimisation

Authors: Makoto Tokuda, Toru Yamamoto

Addresses: Department of Information Engineering, Yuge National College of Maritime Technology, 1000 Shimoyuge-Yuge, Kamijima, Ochi, Ehime, 794-2593, Japan. ' Division of Electrical Systems and Mathematical Engineering, Faculty of Engineering, Hiroshima University, 1-4-1 Kagamiyama, Higashi-Hiroshima, Hiroshima, 739-8527, Japan

Abstract: Most of process systems such as chemical plants are considered as non-linear systems. The global linear approximation for the systems with the strong non-linearities might cause the large modelling error. It is then difficult to obtain the good control performance, even if the controllers are suitably designed based on the models. In this paper, a design method of the PID control system with the data-driven modelling function has been proposed. In the proposed method, local linear models are designed with the multiple datasets selected from the database when needed. Also, the time-variant system parameters are automatically adjusted by using the particle swarm optimisation. Finally, the effectiveness of the proposed method is numerically evaluated through applications to the non-linear systems and the time-variant systems.

Keywords: PID control; nonlinear systems; time-variant systems; particle swarm optimisation; PSO; data-driven modelling; generalised minimum variance control; GMVC; process systems; chemical plants.

DOI: 10.1504/IJMIC.2011.040493

International Journal of Modelling, Identification and Control, 2011 Vol.13 No.1/2, pp.88 - 96

Published online: 31 May 2011 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article