頁籤選單縮合
題 名 | Design of an Optimization Algorithm Inspired by the Bacterial Evolution |
---|---|
作 者 | Lai, Hung-ren; Wong, Ching-chang; | 書刊名 | International Journal of Electrical Engineering |
卷 期 | 12:1 民94.02 |
頁 次 | 頁51-62 |
分類號 | 448.6 |
關鍵詞 | Genetic algorithms; Bacterial evolution; Polynomial estimation problem; Traveling salesman problem; |
語 文 | 英文(English) |
英文摘要 | An optimization algorithm named multiple-stage genetic algorithm (msGA) is proposed in this paper. Several intuitive and realizable mechanisms and operators that imitate from the behaviors and characteristics in the bacterial evolution are implemented in the proposed msGA to increase the search speed and improve the solution accuracy. An encoding method with variable-length parameter structure for the representation of chromosome is proposed to increase the feasibility of msGA for different kinds of optimization issues. Besides, three suggested variable operator: transformation, crossover, and adaptation, are employed to change the length of chromosomes, exchange parts of genes between two chromosomes, and perform a local optimization, respectively. Some numerical optimization examples and comparisons are presented to illustrate the advantage in the search ability of the proposed msGA. The least squares regression polynomial estimation problem and the traveling salesman problem (TSP) are also utilized to illustrate the feasibility of the proposed multiple-stage structure and the purpose-designed variable operators. |
本系統中英文摘要資訊取自各篇刊載內容。