基于进化计算的多目标鲁棒优化方法

发布时间：2018-06-11 22:00

  本文选题：多目标鲁棒优化 + 严格鲁棒性　； 参考：《系统工程与电子技术》2009年05期

【摘要】：工程应用中求解多目标优化问题时,所求的解既要具有较高的质量,又要满足指定的鲁棒性要求。对已有的多目标优化解的鲁棒性度量方法进行了分析,基于用户提出的严格鲁棒性要求,给出了一种严格鲁棒性度量方法并建立了求多目标鲁棒Pareto最优解的数学模型。模型归结为一个嵌套的双重优化过程,外层优化过程用于搜索高质量的解,内层优化过程用于测量候选解的鲁棒性度量指标。以进化计算作为搜索引擎,给出了实施模型的算法,仿真结果表明了方法是有效的。
[Abstract]:When solving multi-objective optimization problem in engineering application, the solution must have high quality and meet the specified requirements of robustness. Based on the strict robustness requirements proposed by users, a strict robustness measurement method is proposed and a mathematical model for finding the optimal solution of multi-objective robust Pareto is established. The model is reduced to a nested dual optimization process in which the outer optimization process is used to search for high quality solutions and the inner layer optimization process is used to measure the robustness of the candidate solution. Using evolutionary computing as the search engine, the algorithm of implementing the model is given. The simulation results show that the method is effective.
【作者单位】： 中南大学信息科学与工程学院;邵阳学院信息工程系;
【基金】：湖南省教育厅科研基金(05C671) 中南大学创新基金(ZB018)资助课题
【分类号】：TP18
，

本文编号：2006846