Particle Swarm Improvement Optimization Algorithm and Performance Study

Abstract:

In this paper, the PSO algorithm is improved, and take mutate of idea into PSO, namely algorithm not set particle boundaries constraints to make them limited in the search interval, but produce the same number of random particle replace those particles of flying away from the search area. Put forward a kind of improved particle swarm optimization algorithm. Use three input XOR problem to improve algorithm testing, the results show that the improved algorithm in the convergence speed and global search ability is superior to the PSO and SGA algorithm, and avoid the prematurity and local convergence.

• Periodical: Advanced Materials Research (Volumes 468-471)
• Pages: 2546-2549
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.468-471.2546
• Online since: February 2012
• Authors: Wei Dong Ji, Ke Qi Wang
• Keywords: Genetic Algorithm (GA), Neural Network (NN), Particle Swarm Optimization Algorithm (PSO)

References

[1] J. Kennedy, R. Eberhart. Particle Swarm Optimization. Proc. IEEE Int. Conf. Neural Networks, (1995), 1942-1948.

[2] GAO Hai-bing, GAO Liang. Electronica Sinica, (2004),32(9),p.1572, In Chinese

[3] Y.W. Leung, Y.P. Wang. IEEE Transactions on Evolutionary Computation, (2001), 5(1),p.41

[4] Y.Shi, R.C. Eberhart. A modified particle swarm optimizer. Proc. IEEE Int. Conf. Evol. Comput., Anchorage, Alaska, May (1998), 69-73.