一种提高多目标进化算法搜索鲁棒最优解效率的方法

A method for improving performance of multi-objective evolutionary algorithms in searching robust optimal solutions

提出将拉丁超立方体抽样用于计算有效目标函数,有效地提高多目标进化算法求解鲁棒最优解的效果;同时提出一种自适应抽样技术,使求解效果和效率都得到了较大的提高。通过与已有方法的对比实验,研究结果表明:本文所提出的方法求解效果好,效率较高。

The performance of robust optimal solutions by using Latin hypercube sampling（LHS） to compute effective objective functions was improved.Furthermore,an adaptive sampling technique was proposed,which can improve the performance and efficiency of multi-objective evolutionary algorithms（MOEAS） at a great level.Through some comparative experiments,the results demonstrate that the methods suggested in this paper are better than the existing approaches both in the performance of robust optimal solutions and the efficiency of MOEAs.