Authors: Jing Jie; Wanliang Wang
Addresses: College of Computer Science and Technology, Zhejiang University of Technology, No. 288 LiuHe Road, Hangzhou 310023, China; Institution of Information and Control Engineering, Zhejiang University of Science and Technology, No. 318 LiuHe Road, Hangzhou 310023, China. ' College of Computer Science and Technology, Zhejiang University of Technology, No. 288 LiuHe Road, Hangzhou 310023, China
Abstract: Inspired by the information prediction existing in the nature intelligent agents, the author has developed a modified particle swarm optimisation (PSO) with a forward PD controller (SPSO-FWPD) earlier. Comprehensive analysis for the model is provided in the paper, including its stabilisation, convergence and dynamic behaviour. Later, another modified PSO with a feedback PD controller (SPSO-FBPD) is presented accompanying some analysis. The introductions of different PD controllers develop the standard PSO (SPSO) with information prediction capability, which can guide the particle to respond to the change of their exemplars correctly and rapidly, and greatly contributes to a successful global search. The proposed methods provide some new ideas for the improvement of SPSO, and are compared with SPSO based on some numerical optimisation simulations. The relative experimental results show SPSO with different PD controller performs better than SPSO on the complex optimisation problems.
Keywords: swarm intelligence; particle swarm optimisation; modified PSO; PD controllers; stabilisation; convergence; dynamic behaviour.
International Journal of Intelligent Information and Database Systems, 2011 Vol.5 No.6, pp.579 - 596
Available online: 17 Oct 2011 *Full-text access for editors Access for subscribers Purchase this article Comment on this article