An Improved VEPSO Algorithm for Multi-objective Optimisation Problems

Kamarul Hawari, Ghazali and Zuwairie, Ibrahim and Faradila, Naim and Kian, Sheng Lim and Salinda, Buyamin and Anita, Ahmad and Sophan Wahyudi, Nawawi and Norrima, Mokhtar (2015) An Improved VEPSO Algorithm for Multi-objective Optimisation Problems. In: The Malaysia-Japan Model on Technology Partnership. Springer, Japan, pp. 253-263. ISBN 978-4-431-54438-8 (print); 978-4-431-54439-5 (online)

[img]
Preview
PDF
fkee-2015-hawari-Improved VEPSO-bk chpter.pdf

Download (34kB) | Preview

Abstract

Multi-objective optimisation problem is the problem which contains more than one objective that needs to be solved simultaneously. The vector evaluated particle swarm optimisation algorithm is widely used for such purpose, where this algorithm optimised one objective using one swarm of particles by the guidance from the best solution found by another swarm. However, this best solution is only updated when a solution is better with respect to the optimised objective and results in poor performance. Therefore, the vector evaluated particle swarm optimisation algorithm is improved by incorporating the non-dominated solutions for guiding the particle movement during optimisation. The performance of the improved algorithm is analysed with several performance measures and simulated on various test functions. The results suggest that the improved algorithm outperformed the performance of the original algorithm.

Item Type: Book Chapter
Uncontrolled Keywords: Electronics and Microelectronics, Instrumentation ; Nanotechnology and Microengineering; Innovation/Technology Management; Engineering Economics, Organization, Logistics, Marketing
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering
Faculty/Division: Faculty of Electrical & Electronic Engineering
Depositing User: Mrs. Neng Sury Sulaiman
Date Deposited: 30 Mar 2015 03:02
Last Modified: 08 Feb 2018 01:40
URI: http://umpir.ump.edu.my/id/eprint/8798
Download Statistic: View Download Statistics

Actions (login required)

View Item View Item