Magnifier particle swarm optimization for numerical optimization

ABSTRACT

A novel particle swarm optimization algorithm based on magnification transformation, called magnifier particle swarm optimization (MPSO), is proposed for the first time in this paper. In the MPSO, we enlarge the range around the best individual of each generation like using a magnifier, while the velocity of particles unchanged. In such a way, MPSO achieves much faster convergence performance and better optimization solving capability than the conventional standard particle swarm optimization and latest clonal PSO by a number of simulations. A detailed description and explanation of the MPSO algorithm are given in the paper. Experiments on fourteen benchmark test functions are conducted and shows the inspiring success that the proposed MPSO speeds up the convergence tremendously, while keeping a good search capability of global solution with much more accuracy. Experiments on fourteen benchmark test functions are conducted to demonstrate that the proposed MPSO algorithm is able to speedup the evolution process distinctly and improve the performance of global optimizer greatly.

REFERENCES

• 1: J. Kennedy and R. Eberhart, "Particle Swarm Optimization," Proceedings of the IEEE International Conference on Neural Networks, Perth, Australia, IEEE Service Center, Vol. 4. Piscataway, NJ, 1995, pp. 1942--1948.
• 2: Y. H. Shi and R. Eberhart, "A Modified Particle Swarm Optimizer," IEEE World Congress on Computational Intelligence, Alaska, ALTEC, vol. 1, 1998, pp.69--73.
• 3: Paul Blenkhorn and David Gareth Evans, "A Screen Magnifier Usin High Level' Implementation Techniques" IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING, vol. 14, no. 4, 2006, pp. 501--504.
• 4: Y. Tan and Z. M. Xiao, "Clonal Particle Swarm Optimization and Its Applications," IEEE Congress on Evolutionary Computation, 2007, pp. 2303--2309.

INDEX TERMS

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

Additional Classification:
  J. Computer Applications
  J.4 SOCIAL AND BEHAVIORAL SCIENCES

General Terms:
Algorithms, Performance

Keywords:
global numerical optimization., particle swarm optimization(pso), swarm intelligence