Particle swarm optimiser with hybrid multi-parent crossover and discrete recombination. (18th October 2011)

• Record Type: Journal Article
• Authors: Jiang, Dazhi; Zeng, Sanyou; Wang, Hui; Wu, Zhijian
• Is Part Of: International journal of intelligent information and database systems. Volume 5:Number 6(2011)
• Page Start: 597
• Page End: 617
• Publication Date: 2011-10-18
• Subjects: particle swarm optimisation -- PSO -- multi-parent crossover -- differential evolution -- discrete recombination -- enhanced evolutionary algorithms; Database management -- Computer programs -- Periodicals; Information retrieval -- Computer programs -- Periodicals; Information storage and retrieval systems -- Computer programs -- Periodicals; Artificial intelligence -- Periodicals; Expert systems (Computer science) -- Periodicals; Intelligent agents (Computer software) -- Periodicals; 006.33
• Languages: English
• ISSNs: 1751-5858

Abstract

Particle swarm optimiser (PSO) has shown good performance in lots of optimisation problems. However, it easily suffers from premature convergence when solving complex optimisation problems. In order to improve the performance of PSO, this paper presents an enhanced evolutionary algorithm named as PSO with hybrid multi-parent crossover and discrete recombination (PSOHMCDR), which is based on the characteristics of PSO, multi-parent crossover algorithm and differential evolution (DE). Experimental results show that PSOHMCDR outperforms other nine algorithms, including six PSO variants and three typical and effective DE variants.