本文選題：多目標(biāo)魯棒優(yōu)化 + 嚴(yán)格魯棒性��； 參考：《系統(tǒng)工程與電子技術(shù)》2009年05期

【摘要】：工程應(yīng)用中求解多目標(biāo)優(yōu)化問(wèn)題時(shí),所求的解既要具有較高的質(zhì)量,又要滿足指定的魯棒性要求。對(duì)已有的多目標(biāo)優(yōu)化解的魯棒性度量方法進(jìn)行了分析,基于用戶提出的嚴(yán)格魯棒性要求,給出了一種嚴(yán)格魯棒性度量方法并建立了求多目標(biāo)魯棒Pareto最優(yōu)解的數(shù)學(xué)模型。模型歸結(jié)為一個(gè)嵌套的雙重優(yōu)化過(guò)程,外層優(yōu)化過(guò)程用于搜索高質(zhì)量的解,內(nèi)層優(yōu)化過(guò)程用于測(cè)量候選解的魯棒性度量指標(biāo)。以進(jìn)化計(jì)算作為搜索引擎,給出了實(shí)施模型的算法,仿真結(jié)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.
【作者單位】： 中南大學(xué)信息科學(xué)與工程學(xué)院;邵陽(yáng)學(xué)院信息工程系;
【基金】：湖南省教育廳科研基金(05C671) 中南大學(xué)創(chuàng)新基金(ZB018)資助課題
【分類號(hào)】：TP18
，

本文編號(hào)：2006846