Particle swarm optimization with information share mechanism

ABSTRACT

This paper proposes an information share mechanism into particle swarm optimization (PSO) in order to use all the useful information of the swarm to prevent premature convergence. The particle in traditional PSO uses only the information from its personal best position and the neighborhood's best position. This mechanism is not with sufficient search information and therefore the algorithm is easy to be trapped into local optima. In the proposed information share PSO (ISPSO), all the particles post their best search information to a share device and any particle can read the information on the device and use the information provided by any other particle to help enhance its search ability. Therefore, the ISPSO can use the whole swarm's information to guide the flying direction. The ISPSO has been applied to optimize multimodal functions, and the experimental results demonstrate that the ISPSO can yield better performance when is compared with the traditional and some other improved PSOs.

REFERENCES

• 1: J. Kennedy and R. C. Eberhart, "Particle swarm optimization," in Proc. IEEE Int. Conf. Neural Networks, Perth, Australia, vol. 4, 1995, pp. 1942--1948.
• 2: J. Kennedy , R. Mendes, Population structure and particle swarm performance, Proceedings of the Evolutionary Computation on 2002. CEC '02. Proceedings of the 2002 Congress, p.1671-1676, May 12-17, 2002
• 3: R. Mendes, J. Kennedy and J. Neves, "The fully informed particle swarm: Simpler, maybe better," IEEE Trans. Evol. Comput., vol. 8, pp. 204--210, Jun. 2004.

INDEX TERMS

Primary Classification:
  I. Computing Methodologies
  I.2 ARTIFICIAL INTELLIGENCE
      I.2.8 Problem Solving, Control Methods, and Search
          Subjects: Heuristic methods

General Terms:
Algorithms, Experimentation, Performance, Reliability

Keywords:
global numerical optimization, information share, particle swarm optimization