An idea of Novel Optimizer using Swarm of Chaotic Dynamical Particles

Yoshikazu Yamanaka; Tsubone Tadashi

Summary

Proceedings of the 2013 International Symposium on Nonlinear Theory and its Applications

2013

Session Number:B2L-B

Session:

Number:240

An idea of Novel Optimizer using Swarm of Chaotic Dynamical Particles

Yoshikazu Yamanaka, Tsubone Tadashi,

pp.240-243

Publication Date:

Online ISSN:2188-5079

DOI:10.15248/proc.2.240

PDF download (486.6KB)

Summary:

In this paper we propose a novel Optimization method called Optimizer using Swarm of Chaotic Dynamical Particles. Our proposed method is based on particles which follow chaotic dynamics. The chaotic dynamics makes the particle complex motion by stretching and holding mechanism even tough the dynamics does not contain any stochastic elements. A swarm consists of the particles which share the information of their own position which is candidate solution like Particle Swarm Optimization(PSO). By the particle's complex behavior and sharing information mechanism, our proposed method has ability to search optimum solution. Our proposed method does not contain any stochastic elements and has only 3 system parameters, therefor, it is easy to implement the algorithm, to analyze particle dynamics and to design system parameters. We compared our proposed method with other deterministic PSO and conventional PSO for well-known benchmark functions and proposed method shown better search ability in almost all of situation.

References:

[1] J. Kennedy and R. Eberhart, “Particle Swarm Optimization,” in Proc. IEEE Int. Conf. Neural Networks, pp. 1942-1948, 1995.

[2] M. Clerc and J. Kennedy, “The particle swarm - explosion, stability, and convergence in a multidimensional complex space,” IEEE Trans. Evol. Comput. , Vol. 6, No. 1, pp. 58-73, 2002.

[3] Takuya Shindo and Kenya Jin'no, “On a Reacceleration Mechanism for Particle Swarm Optimizer,” Journal of Signal Processing, Vol. 15, No. 6, pp. 407-516, November 2011.

[4] R. C. Eberhart and Y. Shi, “Comparing inertia weights and constriction factors in particle swarm optimization,” in Proc. 2000 Congr. Evol. Comput. , San Diego, CA, July 2000, pp. 84-88.