A Hybrid Particle Swarm Algorithm for Function Optimization

Particle swarm optimization (PSO) is one of the evolutionary techniques based on swarm intelligence, which has show good performance in many optimization problems. This paper proposes a new learning strategy to help particles learn experiences from other previous best particles. In order to verify the proposed approach (HPSO), this paper investigates the effects of learning factor on six well-known benchmark functions. Additionally, comparison of HPSO with standard PSO and comprehensive learning PSO shows that HPSO outperforms them on most test functions. Keywords-particle swarm optimization; learning strategy; function optimization

[1]  Yue Shi,et al.  A modified particle swarm optimizer , 1998, 1998 IEEE International Conference on Evolutionary Computation Proceedings. IEEE World Congress on Computational Intelligence (Cat. No.98TH8360).

[2]  Zhihua Cui,et al.  Particle swarm optimization with FUSS and RWS for high dimensional functions , 2008, Appl. Math. Comput..

[3]  L. Coelho,et al.  Predictive Controller Tuning Using Modified Particle Swarm Optimization Based on Cauchy and Gaussian Distributions , 2005 .

[4]  James Kennedy,et al.  Particle swarm optimization , 2002, Proceedings of ICNN'95 - International Conference on Neural Networks.

[5]  Hui Wang,et al.  Opposition-based particle swarm algorithm with cauchy mutation , 2007, 2007 IEEE Congress on Evolutionary Computation.

[6]  Renato A. Krohling,et al.  Gaussian particle swarm with jumps , 2005, 2005 IEEE Congress on Evolutionary Computation.

[7]  Chongzhao Han,et al.  Knowledge-based cooperative particle swarm optimization , 2008, Appl. Math. Comput..

[8]  Jing J. Liang,et al.  Evaluation of Comprehensive Learning Particle Swarm Optimizer , 2004, ICONIP.

[9]  Wenbo Xu,et al.  An improved quantum-behaved particle swarm optimization algorithm with weighted mean best position , 2008, Appl. Math. Comput..