一种新的遗传算法求解约束优化问题

A New Genetic Algorithm for Constrained Optimization

采用十进制编码的遗传算法,在进化计算中提出用排序、保留优秀个体、智能变异和随机变异代替以前遗传算法中的选择、交叉和变异。此算法具有简单的计算原理,简化了遗传算法的编程;利用变换后的目标函数构造适配值,能较快的找到最优解。数值实验表明该算法在求解有约束优化问题上表现良好。

Decimal coding is adopted in the textual arithmetic,using ranking,elitism,random and intelligent mutation replace selection, crossover and mutation in GA The method possessed simple computing basis and simpler programming,and if we make use of object function transformed to construct fitness evaluation we can find the best solution to the problem quickly Numerical experiments indicate that the performance of genetic algorithm on constrained optimization is sound