High-Dimension Optimization Problems Using Specified Particle Swarm Optimization
Particle Swarm Optimization (PSO), proposed by Dr. J. Kennedy and Professor R. Eberhart in 1995, attracts many attentions to solve for a lot of real uni-modal/multi-modal optimization problems. Due to its simplicity of parametersetting and computational efficiency, PSO becomes one of the most popular algorithms for optimization search. Since 1995, many researchers provide different algorithms to set parameters for convergence, explosion and exploitation potential of PSO. Most of the proposed methods are to find a general PSO (called Standard PSO, SPSO) for most of the benchmark problems. However, those may not be suitable to a specified problem, for example, Shaffer or Rosenbrock problems; especially the dimension of the problem is high. On the contrary, with to the difficult problem such as, Rosenbrock, a more proper specified PSO is needed for this high-dimension problem. Therefore, for each problem after more understanding the characteristic of the problem, a SPecified PSO (SPPSO) is proposed. Apply this idea to 5 benchmark problems, such as sphere, quatric, Rosenbrock, Griewank, and Rastrigin functions, four different SPPSO algorithms are proposed with good results in the end.
KeywordsGenetic algorithms particle swarm optimization mutation optimization
Unable to display preview. Download preview PDF.
- 1.Kennedy, J., Eberhart, R.: Particle Swarm Optimization. In: Proceedings of the IEEE International Joint Conference on Neural Networks, pp. 1942–1948 (1995)Google Scholar
- 3.Shi, Y., Eberhart, R.: A Modified Particle Swarm Optimizer. In: The 1998 IEEE International Conference on Evolutionary Computation Proceedings, IEEE World Congress on Computational Intelligence, pp. 69–73 (1998)Google Scholar
- 4.Cai, X., Cui, Z., Zeng, J., Tan, Y.: Individual Parameter Selection Strategy for Particle Swarm Optimization: Particle Swarm Optimization. InTech Education and Publishing (2009)Google Scholar
- 5.Vesterstrom, J., Thomsen, R.: A Comparative Study of Differential Evolution, Particle Swarm Optimization, and Evolutionary Algorithms on Numerical Benchmark Problems. In: Proc. IEEE Congress Evolutionary Computation, Portland, OR, June 20-23, pp. 1980–1987 (2004)Google Scholar
- 6.Chou, P.: Principles and Applications of Genetic Algorithms. ChiamHwa Book Company, Taipei (2006) (in Chinese)Google Scholar
- 7.Chou, P.: Improved Particle Swarm Optimization with Mutation. In: MSI-2009 IASTED International Conference, Beijing, China (2009)Google Scholar
- 8.Chou, P.: A Proposal for Improved Particle Swarm Intelligence. In: IEEE-3CA International Conference 2010, Tainan, Taiwan (2010)Google Scholar
- 10.Chou, P., Lian, J.: Enforced Mutation to Enhancing the Capability of Particle Swarm Optimization Algorithms. In: The Second International Conference on Swarm Intelligence, Chongqiang, China, pp. 12–15 (June 2011)Google Scholar
- 11.Hatanaka, T., Korenaga, T., Kondo, N., Uosaki, N.: Search Performance for PSO in High Dimensional Spaces Particle Swarm Optimization: InTech open (2009)Google Scholar
- 12.Clerc, M.: A Mini-benchmark, http://clercmaurice.free.fr/pso/
- 13.Yeh, Y.: A Simple Hybrid Particle Swarm Optimization: Advances in Evolutionary Algorithms. I-Tech Education and Publishing, Vienna (2008)Google Scholar