A New Genetic Algorithm Based on Optimal Solution Orientation

Abstract:

Many researchers have indicated that standard genetic algorithm suffers from the dilemma---premature or non-convergence. Most researchers focused on finding better search strategies, and designing various new heuristic methods. It seemed effective. From another view, we can transform search space with a samestate-mapping. A special genetic algorithm applied to the new search space would achieve better performance. Thus, we present a new genetic algorithm based on optimal solution orientation. In this paper, a new genetic algorithm based on optimum solution orientation is presented. The algorithm is divided into "optimum solution orientation" phase and "highly accurately searching in local domain of global optimal solution" phase. Theoretical analysis and experiments indicate that OSOGA can find the "optimal" sub domain effectively. Cooperating with local search algorithm, OSOGA can achieve highly precision solution with limited computing resources.

• Periodical: Advanced Materials Research (Volumes 139-141)
• Pages: 1779-1784
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.139-141.1779
• Online since: October 2010
• Authors: Quan Wang, Jin Chao Liu, Pan Wang, Juan Ying Qin
• Keywords: Genetic Algorithm (GA), Optimal, Solution Orientation

References

[1] P. Wang and X. G Li . Study on Multi-Person and Multi-Objective Decision-Making Method of Hybrid Objective Systems. Systems Engineer and Electronics 1999. No 12. vol(21).

[2] F. Zhun, S. Kisung, Ronald C. Rosenberg, J.J. Hu, Erik D. Goodman. Computational Synthesis of Multi-Domain Systems. The 2003 AAAI Spring Symposium Computational Synthesis: From Basic Building Blocks to High Level Functionality, pages 59-66, Stanford, California, (2003).

[3] J.C. Wang, F. Zhun, Janis P. Terpenny and Erik D. Goodman. Knowledge Interaction With Genetic Programming in Mechatronic Systems Design Using Bond Graphs. IEEE transactions on system, man, and cybernetics-part C: applications and reviews. vol. 35, No. 2 May (2005).

DOI: 10.1109/tsmcc.2004.841915

[4] P. Wang, J.K. Wa, S. Feng. Multi-population co-evolutionary algorithm based on the divide and conquer, principle. Systems Engineering and Electronics, Vol. 26., No. 11.

[5] P. Wang, S. Feng and W. Wei. Study on Modular Mechanism of the Co-Evolution Genetic Algorithm. Journal of WHUT (Information & mana gemenet engineering) Vol. 25 No. 6 Dec. (2003).

[6] Y. Liu,X. Yao, and Tetsuya Higuchi. Evolutionary Ensembles with Negative Correlation Learning. IEEE transactions on evolutionary computaiton, Vol. 4, No. 4, November (2000).

DOI: 10.1109/4235.887237

[7] Hiroshi Sato, Akira Namatame Co-evolution In Social Interactions International Journal of Computational Intelligence and Applications Vol. 2, No. 1. (2002).