A New Combined Particle Swarm Optimization Algorithm Based Golden Section Strategy

Abstract:

Based the defects of global optimal model falling into local optimum easily and local model with slow convergence speed during traditional PSO algorithm solving a complex high-dimensional and multi-peak function, a two sub-swarms particle optimization algorithm is proposed. All particles are divided into two equivalent parts. One part particles adopts global evolution model, while the other part uses local evolution model. If the global optimal fitness of the whole population stagnates for some iteration, a golden rule is introduced into local evolution model. This strategy can substitute the partial perfect particles of local evolution for the equivalent worse particles of global evolution model. So, some particles with advantage are joined into the whole population to make the algorithm keep active all the time. Compared with classic PSO and PSO-GL(A dynamic global and local combined particle swarm optimization algorithm, PSO-GL), the results show that the proposed PSO in this paper can get more effective performance over the other two algorithm in the simulation experiment for four benchmark testing function.

• Periodical: Advanced Materials Research (Volumes 308-310)
• Pages: 1099-1105
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.308-310.1099
• Online since: August 2011
• Authors: Hui Fan
• Keywords: Golden Rule Diversity, Model, Optimization, Particle Sawrm Optimization Algorithm

References

[1] Kennedy J, Eberhart R C. Particle swarm optimization [A]. Proceeding of the IEEE International Conference on Neural Networks[C]. Piscataway, NJ: IEEE Service Center, 1995: 1942-1948.

[2] Eberhart R. C, Kennedy J. A new optimizer using particle swarm theory[A]. Proceeding of the Sixth International Symposium on Micro Machine and Human Science[C]. Nagoya,Japan: IEEE Service Center, 1995: 39 - 43.

DOI: 10.1109/mhs.1995.494215

[3] X. H. Shi, Y. C. Lianga, H.P. Lee. Particle swarm optimization-based algorithms for TSP and generalized TSP [J]. Information Processing Letters. 2007,103(5): 169- 176.

DOI: 10.1016/j.ipl.2007.03.010

[4] Ruey-Maw Chen a, Chung-Lun Wu, Chuin-Mu Wang, etc. Using novel particle swarm optimization scheme to solve resource-constrained scheduling problem in PSPLIB[J]. Expert Systems with Applications. 2010,37(3): 1899-1910

DOI: 10.1016/j.eswa.2009.07.024

[5] Hamidreza Modares, Alireza Alfi , Mohammad-Mehdi Fateh. Parameter identification of chaotic dynamic systems through an improved particle swarm optimization[J]. Expert Systems with Applications. 2010,37(5): 3714-3720.

DOI: 10.1016/j.eswa.2009.11.054

[6] Tsung-Lieh Lin, Shi-Jinn Horng, Tzong-Wann Kao, etc. An efficient job-shop scheduling algorithm based on particle swarm optimization[J]. Expert Systems with Applications. 2010,37(3): 2629-2636.

DOI: 10.1016/j.eswa.2009.08.015

[7] Bin Jiao , Zhigang Lian, Qunxian Chen. A dynamic global and local combined particle swarm optimization algorithm[J]. Chaos, Solitons and Fractals, 2009(42): 2688–2695.

DOI: 10.1016/j.chaos.2009.03.175