基于偏相关性分析的多设计要求快速优化设计方法
Fast optimization design with multi-requirements based on the analysis of partial correlation
-
摘要: 针对飞行器气动外形优化设计中的多设计点、多目标、多工程约束等多设计要求问题，提出了一种新优化设计方法——多设计要求快速优化设计方法。该方法立足于统计学中的偏相关性分析和线性回归理论，通过对设计变量和设计要求进行偏相关性分析，采用线性回归的方法构建设计变量与设计要求之间的转换模型，将多个设计要求转化为对设计变量的约束，从而简化了优化设计模型，减少了优化过程中调用求解器次数，提高了优化效率。利用该方法对RAE 2822和HSNLF(1)-0213翼型进行了多设计要求优化设计算例验证，并将优化结果与Pareto多目标优化方法结果进行对比，结果证明了本文方法的有效性和可靠性。Abstract: Aiming at aircraft aerodynamic optimization design problems with multi-requirements including multi-designing points, multi-objects and multi-constraints, a new method for optimization design is proposed in this paper: fast optimization design with multi-requirements. The method is established on the partial correlation and linear regression theories which originated from statistics. According to the analysis of partial correlation between design variables and design requirements, some of the requirements are transformed into the constraints of variables through the transformation model established by the linear regression theory, and the optimization result shows that the simpler the optimization model is, the shorter time will be costed in the optimization design. Two multi-requirements optimization design examples have been investigated in this paper, in which the RAE2822 and HSNLF(1)-0213 are as the initial foils respectively, and the optimization result by the method in this paper are compared with that of the Pareto multi-object method, which verifies the validity and reliability of the method in this paper.
-
-
[1] EYI S, LEE K D. Inverse airfoil design using the Navier-Stokes equations[R]. AIAA 1993-0972. [2] KIM H J, RHO O H. Aerodynamic design of transonic wing using the target pressure optimization approach[R]. AIAA 1998-0599. [3] 杨希祥, 李晓斌, 肖飞. 智能优化算法及其在飞行器优化设计领域的应用综述[J]. 宇航学报, 2009, 11(30). [4] 林锉云, 董加礼. 多目标优化的方法和理论[M]. 吉林教育出版社, 1992. [5] 吴卫东, 黄洪钟, 李海滨. 对称模糊优化模型评价函数解法[J]. 大连理工大学学报, 2003, 43(3). [6] CHRISTOPHER A M, ACHILLE M. Development of a Pareto-based concept selection method[R]. AIAA 2002-1231. [7] 夏露, 高正红, 李天. 飞行器外形多目标多学科综合优化设计方法研究[J]. 空气动力学学报, 2003, 21(3): 275-281. [8] 夏露. 飞行器外形气动、隐身综合优化设计方法研究[D]. [博士学位论文]. 西安: 西北工业大学, 2004. [9] 郝黎仁, 等. SPSS实用统计分析[M]. 中国水利水电出版社.
[10] 阮桂海, 蔡建琼, 朱志海. 统计分析应用教程[M]. 北京: 清华大学出版社, 2003,7.
[11] YANG Q, HU B. Using matrix to fifure random vector′s correlative degree[J]. Journal of Wuhan University of Technology, 2002, 6: 26-3.
[12] 薛薇. 统计分析与SPSS的应用[M]. 中国人民大学出版社, 2001, 11.[10] 姬景周. 统计学原理[M]. 中国商业出版社, 1997. [11] 冯启磊, 曾文艺. 模糊最小一乘线性回归[J]. 北京师范大学学报(自然科学版), 2008, 44(4). [12] 孙颉, 吕震宙. 模糊线性回归及其在模糊可靠性分析中的应用[J]. 西北工业大学学报, 2006, 24(1).
[16] RAYMOND M H, PRESTON A H. Wing design by numerical optimization[J]. Journal of Aircraft, 1978, 15(7): 407-412.
计量
- 文章访问数: 261
- HTML全文浏览量: 7
- PDF下载量: 1028
下载:
