基于改进实数遗传算法的函数全局优化

将一种改进的实数遗传算法用于函数全局优化。改进的算法建立在对基本实数遗传算法搜索特性判断的基础上。文中对实数遗传算法的基本操作进行了简单的讨论和选择 ,将一种混沌序列作为刺激因素加入到算法中 ,并将区域划分与取舍的思想应用到算法结构改进中。数值实验显示 ,新方法对寻找复杂问题的全局解、提高搜索精度方面较基本实数遗传算法有较大改进。

OPTIMIZATION BASED ON IMPROVED REAL-CODED GENETIC ALGORITHM

An improved real coded genetic algorithm is proposed for global optimization of functions. The new algorithm is based on the judgement of the searching performance of basic real coded genetic algorithm. The operations of basic real coded genetic algorithm are briefly discussed and selected. A kind of chaos sequence is described in detail and added in the new algorithm as a disturbance factor. The strategy of field partition is also used to improve the structure of the new algorithm. Numerical experiment shows that the new genetic algorithm can find the global optimum of complex functions with satistaiting precision.