基于设计变量空间范围自适应优化设计方法研究

摘要

Based on a study of optimization design method and distribution of design variables space,an adaptive optimal design method of the design variable space is established,and the problem about design variable spatial extent choice is solved.In this method,the conception of aggregation is built,which is used to describe the statistical distribution rule of the design variables spaces.This rule is used to adjust the design variables spaces automatically and to reconstruct the design variable during the optimization.The design variable space extent is adapted during the optimization by this method.Optimal solutions for requirements are got in a larger spatial scale,at the same time the efficiency is improved by reconstruction of design variables.In this paper,the optimal design problems of NACA 0012 and NLF(1) 0416 airfoils are solved by the adaptive optimal design method of the design variable space,the optimal design results demonstrate the optimal design method is feasible.At the same time,in comparison with the stable optimal design method of the design variable space,the results demonstrate that this optimization method can search the optimal solutions in a larger spatial scale and in a higher ability.%通过对翼型优化设计方法的研究,对设计变量在设计空间的分布进行了探讨,建立了设计变量空间范围自适应优化设计方法,解决了优化设计模型建立过程中的设计变量空间范围的选择问题.该方法建立了聚集度的概念,利用聚集度对优化设计过程中设计变量在设计空间内的分布规律进行统计分析,并根据统计分布规律对设计变量的设计空间进行调整,将设计变量重构,实现了优化设计过程中的设计变量空间自适应.该设计变量空间范围自适应优化设计方法,一方面能够在给定的条件下扩大搜索范围来搜索满足工程设计需要的最优解,一方面又能够通过对设计变量的重构提高搜索算法的搜索效率.利用该方法对NACA 0012和NLF(1) 0416翼型进行了优化设计,并与固定设计变量空间范围的优化设计方法进行对比分析.优化结果验证了本文提出的优化设计方法具有一定的可行性,可以在更大范围内找到最优解,并具有较高的优化搜索效率.