頁籤選單縮合
題 名 | 適應性突變運算及其運用=Adaptive Mutation Operators and Its Applications |
---|---|
作 者 | 陳木松; 廖鴻翰; | 書刊名 | 大葉學報 |
卷 期 | 7:1 1998.12[民87.12] |
頁 次 | 頁91-101 |
分類號 | 448.595 |
關鍵詞 | 基因演算法; 適應性突變運算; 突變機率; Genetic algorithms; Adaptive mutation operators; Mutation probability; |
語 文 | 中文(Chinese) |
中文摘要 | 基因演算法是以『自然選擇』及『適者生存』為主的全域搜尋法則,由於這一特性使得基因演算法特別適用於高維度、多極點、或不連續參數空間的極值搜尋問題。基因演算法的演化能力決定於其運算子的效率,以導引系統至全域的最佳極點。本文的研究主要是提出適應性突變運算的方法,使基因演算法兼具有區域與全域搜尋的能力。此外突變機率也將依個體的適應值動態的改變。所以整合區域性搜尋的方法,配合基因演算法本身具有全域搜尋的能力,將可提高其搜尋全域最佳極值的能力,並加速基因演算法的收斂性。最後本文將以10個常用的複雜函數測試所提方法的效率。 |
英文摘要 | Genetic Algorithms (GAs) are global optimization techniques based on "natural selection" and "survival of the fittest" principles that can be applied in high dimensional, multi-modal, and complex problems. The performance of GAs relies on efficient search operators to guide the system towards global optima. In this paper, we proposed adaptive mutation operators to refine the solutions such that GAs possess both the global and local search capabilities. Moreover, the mutation probability is adaptive with respect to the fitness of each individual. The integration of local optimization procedure with the exploration property of GAs enhances the abilities of GAs in searching global optima as well as in speeding convergence. Finally, 10 benchmark functions are simulated to show the superiority of the proposed methods. |
本系統中英文摘要資訊取自各篇刊載內容。