Title: Bacterial foraging algorithm-based optimisation for controlling conditioner temperature of a ring die granulator

Authors: Kun Zhang; Peijian Zhang; Jianguo Wu; Hossein Farid Ghassem Nia; Huiyu Zhou

Addresses: School of Mechatronic Engineering and Automation, Shanghai University, Shanghai 200072, China ' School of Electrical Engineering, Nantong University, Nantong, Jiangsu 226019, China ' School of Electrical Engineering, Nantong University, Nantong, Jiangsu 226019, China ' School of Computer Science and Electronic Engineering, University of Essex, Colchester CO4 3SQ, UK ' School of Electronics, Electrical Engineering and Computer Science, Queen's University Belfast, Belfast, BT3 9DT, UK

Abstract: Support vector regression (SVR) is firmly grounded in the framework of statistical learning theory that has been deployed to solve many engineering problems in recent years. To optimise its parameters and achieve optimised results, this paper deploys a novel bacterial foraging algorithm (BFA), combining with support vector machine (SVM) to achieve the good condition temperature prediction of a ring die granulator. With a strong globally searching capability, the proposed method can achieve dynamic optimisation of systematic parameters and overcome the problem of inefficiency in selecting optimal parameters. Simulation results show that the proposed bacterial foraging algorithm is effective in parameter optimisation of the controller for a ring die granulator.

Keywords: bio-inspired optimisation; bacterial foraging algorithm; BFA; support vector machines; SVM; prediction model; ring die granulator; temperature control; conditioner temperature; simulation; parameter optimisation.

DOI: 10.1504/IJMIC.2014.066267

International Journal of Modelling, Identification and Control, 2014 Vol.22 No.4, pp.357 - 365

Published online: 27 Dec 2014 *

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