A novel improved accelerated particle swarm optimization algorithm for global numerical optimization. Issue 7 (30th September 2014)

• Record Type: Journal Article
• Authors: Wang, Gai-Ge; Hossein Gandomi, Amir; Yang, Xin-She; Hossein Alavi, Amir
• Is Part Of: Engineering computations. Volume 31:Issue 7(2014)
• Page Start: 1198
• Page End: 1220
• Publication Date: 2014-09-30
• Subjects: Computer-aided engineering -- Periodicals; Computer graphics -- Periodicals; 620.00285
• DOI: 10.1108/EC-10-2012-0232
• Languages: English
• ISSNs: 0264-4401

Abstract

<abstract> <title> <x content-type="archive" xml:space="preserve">Abstract</x> </title> <sec> <title content-type="abstract-heading">Purpose</title> <p> – Meta-heuristic algorithms are efficient in achieving the optimal solution for engineering problems. Hybridization of different algorithms may enhance the quality of the solutions and improve the efficiency of the algorithms. The purpose of this paper is to propose a novel, robust hybrid meta-heuristic optimization approach by adding differential evolution (DE) mutation operator to the accelerated particle swarm optimization (APSO) algorithm to solve numerical optimization problems. </p> </sec> <sec> <title content-type="abstract-heading">Design/methodology/approach</title> <p> – The improvement includes the addition of DE mutation operator to the APSO updating equations so as to speed up convergence. </p> </sec> <sec> <title content-type="abstract-heading">Findings</title> <p> – A new optimization method is proposed by introducing DE-type mutation into APSO, and the hybrid algorithm is called differential evolution accelerated particle swarm optimization (DPSO). The difference between DPSO and APSO is that the mutation operator is employed to fine-tune the newly generated solution for each particle, rather than random walks used in APSO. </p> </sec> <sec> <title content-type="abstract-heading">Originality/value</title> <p> – A novel hybrid method is proposed and used to optimize 51 functions. It is compared with other methods to show its effectiveness. The effect of the DPSO parameters on convergence and performance is also studied and analyzed by detailed parameter sensitivity studies.</p> </sec> </abstract>