Author:
Zemzami Maria,El Hami Norelislam,Itmi Mhamed,Hmina Nabil
Abstract
Meta-heuristic PSO has limits, such as premature convergence and high running time, especially for complex optimization problems. In this paper, a description of three parallel models based on the PSO algorithm is developed, on the basis of combining two concepts: parallelism and neighborhood, which are designed according to three different approaches in order to avoid the two disadvantages of the PSO algorithm. The third model, SPM (Spherical-neighborhood Parallel Model), is designed to improve the obtained results from the two parallel NPM (Neighborhood Parallel Model) and MPM (Multi-PSO Parallel Model) models. The experimental results presented in this paper show that SPM model performed much better than both NPM and MPM models in terms of computing time and solution quality.
Subject
Control and Optimization,Modeling and Simulation
Reference14 articles.
1. Kennedy J., Eberhart R., Particle swarm optimization, in Proceedings of the IEEE International Joint Conference on Neural Networks, IEEE Press, vol. 8, 1995, pp. 1943–1948
2. Kennedy J., Mendes R., Population structure and particle swarm performance, in Proceedings of the IEEE 2002 Congress on Evolutionary Computation, 2002
3. Poli R., Langdon W.B., Holland O., Extending particle swarm optimization via genetic programming, in: Keijzer M., Tettamanzi A.G.B., Collet P., van Hemert J., Tomassini M. (Eds.), Genetic Programming, LNCS (Springer, Heidelberg, 2005), vol. 3447, pp. 291–300
4. A particle swarm optimization approach for permutation flow shop scheduling problem
Cited by
7 articles.
订阅此论文施引文献
订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献