Genetic Algorithm Using the Bimodal Operation to Prevent Prematurity and Reduce Computational Time

Abstract:

A probability associated with population fitness is theoretically derived and used as a parameter in the Genetic Algorithm (GA) to switch the operation modes between the population regeneration, and the genetic operation of crossover and mutation in order to effectively prevent the prematurity and reduce the number of function evaluations. The proposed genetic algorithm using the bimodal operation is employed to search for the global optima of five objective functions, whose results are compared to the conventional GAs using a single operation mode. The simulation results demonstrate that the proposed GA can achieve a better convergence performance than the conventional ones when multimodal objective functions are searched for their global optima.

• Periodical: Applied Mechanics and Materials (Volumes 182-183)
• Pages: 1987-1991
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.182-183.1987
• Online since: June 2012
• Authors: Ling Tzu Tseng
• Keywords: Genetic Algorithm (GA), Global Optima, Objective Functions, Prematurity

References

[1] I. Eksin and O. K. Erol: Evolutionary algorithm with modifications in the reproduction phase, vol. 148, issue 2, pp.75-80 (IEE Proceedings - Software 2001).

DOI: 10.1049/ip-sen:20010503

[2] M. N. Howell, T. J. Gordon and F. V. Brandao: Genetic learning automata for function optimization, vol. 32, issue 6, p.804–815 (IEEE Transactions on Systems, Man and Cybernetics, Part B 2002).

DOI: 10.1109/tsmcb.2002.1049614

[3] P. S. Georgilakis, N. D. Doulamis, A. D. Doulamis, N. D. Hatziargyriou and S.D. Kollias: A novel iron loss reduction technique for distribution transformers based on a combined genetic algorithm - neural network approach, vol. 31, issue 1, pp.16-34.

DOI: 10.1109/5326.923265

[4] Information on www. systemtechnik. tu-ilmenau. de/~pohlheim/GA_Toolbox/fcnindex. html.