基于多重耦合聚类的复杂产品多变量关联设计模型分解  

Multivariable Correlative Model Decomposition for Complex Product Design Based on Clustering with Multiple Couplings

作　　者：彭翔[1] 刘振宇[1] 谭建荣[1] 卜王辉[1] 

机构地区：[1]浙江大学CAD&CG国家重点实验室,杭州310027

出　　处：《机械工程学报》2013年第3期111-121,共11页Journal of Mechanical Engineering

基　　金：国家重点基础研究发展计划(973计划;2011CB706503);国家自然科学基金(51075357);高等学校博士点专项科研基金(20090101110040)资助项目

摘　　要：为了提高复杂产品设计模型分解后的聚合度,提高复杂产品优化设计效率,提出一种基于多重耦合聚类的复杂产品多变量关联设计模型分解方法。基于敏感度计算和耦合特性分析,建立表征设计模型中设计变量与设计函数耦合关联、设计变量耦合关联、设计函数耦合关联等多重耦合强弱的综合耦合度矩阵。然后通过综合耦合度矩阵降维和耦合二元树构建,实现耦合度大的变量函数初步聚类;通过二元树分支间耦合度比较,实现变量函数集聚类。最后以二元树分支为可选分割点,以分解后模型聚合度最高为目标实现设计模型分解。以大型空分装备中填料设计模型分解和透平膨胀机设计模型分解为例验证了方法的有效性。To raise the clustering degree after design decomposition, and then increase design efficiency of complex products, a design decomposition method based on clustering with multiple couplings is proposed. The matrix of synthetic coupling degrees （MSCD）, which includes couplings between variables and functions, couplings among variables, and couplings among functions, is created based on sensitivity analysis and couplings characteristics analysis to reflect amounts of coupling degrees. Initial clustering of variables and functions are accomplished through dimensional reduction of MSCD and structure of coupling binary tree. Clustering of aggregations of variables and functions are accomplished through comparison of coupling degrees among branches of the binary tree. The design model is decomposed to reach the greatest clustering degrees. The packing and turbo expander in large scale air separation equipment are taken as examples to verify the effectiveness of the approach.

关 键 词：多重耦合 综合耦合度矩阵 二元树 聚合度 设计模型分解 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]