Adaptive Particle Swarm Optimization for Continuous Domain

Abstract:

An Adaptive particle swarm optimization algorithm is proposed. Algorithm combines with pareto local search (PLS) method and adaptively adjusts flying time. During the search process, our algorithm can enhance the local search ability of particle swarm optimization (PSO) thought adding random perturbation to local search. The flying time of every particle in our algorithm can adaptively adjust with the evolutionary generations. Some optimization tests of the standard benchmark function confirm that our method has a stronger ability of global optimization.

• Periodical: Advanced Materials Research (Volumes 204-210)
• Pages: 1139-1142
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.204-210.1139
• Online since: February 2011
• Authors: Cheng Ming Qi
• Keywords: Continuous Domain, Pareto Local Search, Particle Swarm Optimization Algorithm (PSO)

References

[1] Kennedy J, Eberhart R C. Particle Swarm Optimization [A]. Proc of the IEEE Intl. Conf. on Neural Networks[C], pages 1942—1948. (1995).

[2] Kennedy J, Eberhart R C. Swarm Intelligence. Morgan Kaufmann Publishers, San Francisco, California, (2001).

[3] Parsopoulos, K. and Vrahatis, M. Recent approaches to global optimization. Natural Computing, 1: 235–306. (2002).

[4] Meng T., Ray T. and Dhar P. Supplementary material on parameter estimation using particle swarm. Preprint submitted to Elsevier Science. (2004).

[5] Kennedy J. The Behavior of Particles. Evolutionary Programming VII, pages 581-587, (1998).

[6] Luis Paquete, Thomas Stützle. A study of stochastic local search algorithms for the biobjective QAP with correlated flow matrices, (2004).

DOI: 10.1016/j.ejor.2004.08.024

[7] Shi Y H, Eberhart R C. Empirical study of particle swarm optimization [A]. In: Proceedings of t he IEEE Congress on Evolutionary Computation [C], Piscataway, NJ, IEEE Service Center, pages 1945～1950. (1999).