Citation

Senthil Arumugam, M. and Rao, M. V. C. and Chandramohan, Aarthi (2008) A new and improved version of particle swarm optimization algorithm with global–local best parameters. Knowledge and Information Systems, 16 (3). pp. 331-357. ISSN 0219-1377

PDF (A new and improved version of particle swarm optimization algorithm with global-local best parameters) 735.pdf Restricted to Repository staff only Download (0B)

Abstract

This paper presents a new and improved version of particle swarm optimization algorithm (PSO) combining the global best and local best model, termed GLBest-PSO. The GLBest-PSO incorporates global-local best inertia weight (GLBest IW) with global-local best acceleration coefficient (GLBest Ac). The velocity equation of the GLBest-PSO is also simplified. The ability of the GLBest-PSO is tested with a set of bench mark problems and the results are compared with those obtained through conventional PSO (cPSO), which uses time varying inertia weight (TVIW) and acceleration coefficient (TVAC). Fine tuning variants such as mutation, cross-over and RMS variants are also included with both cPSO and GLBest-PSO to improve the performance. The simulation results clearly elucidate the advantage of the fine tuning variants, which sharpen the convergence and tune to the best solution for both cPSO and GLBest-PSO. To compare and verify the validity and effectiveness of the GLBest-PSO, a number of statistical analyses are carried out. It is also observed that the convergence speed of GLBest-PSO is considerably higher than cPSO. All the results clearly demonstrate the superiority of the GLBest-PSO.

Item Type:	Article
Subjects:	T Technology > T Technology (General) Q Science > QA Mathematics > QA71-90 Instruments and machines > QA75.5-76.95 Electronic computers. Computer science
Divisions:	Faculty of Engineering and Technology (FET)
Depositing User:	Ms Suzilawati Abu Samah
Date Deposited:	09 Sep 2011 03:32
Last Modified:	09 Sep 2011 03:32
URII:	http://shdl.mmu.edu.my/id/eprint/2278

Downloads

Edit (login required)