中国机械工程 ›› 2010, Vol. 21 ›› Issue (06): 668-671,689.

• 信息技术 • 上一篇    下一篇

基于分级自适应技术车身结构多参数大规模问题快速计算方法研究

雷飞;韩旭
  

  1. 湖南大学汽车车身先进设计制造国家重点实验室,长沙,410082
  • 出版日期:2010-03-25 发布日期:2010-04-02
  • 基金资助:
    国家杰出青年基金资助项目(10725208);高等学校博士学科点专项科研基金资助项目(20070532021)
    National Funds for Distinguished Young Scholars( No. 10725208);
    Specialized Research Fund for the Doctoral Program of Higher Education of China(No. 20070532021)

A Study of Rapid Evaluation for Structural Behavior of Multi-parameterized and Large-scale Problem in Vehicle Body Design

Lei Fei;Han Xu
  

  1. State Key Laboratory of Advanced Design and Manufacturing for Vehicle Body,Hunan niversity,Changsha,410082
  • Online:2010-03-25 Published:2010-04-02
  • Supported by:
     
    National Funds for Distinguished Young Scholars( No. 10725208);
    Specialized Research Fund for the Doctoral Program of Higher Education of China(No. 20070532021)

摘要:

提出一种基于分级自适应技术的车身结构多参数大规模问题快速计算方法。该方法可以在保证数值稳定性的基础上得到较高的计算精度;在单元水平上对结构的刚度矩阵进行参数化分解,得到显式表达的参数化有限元格式;将该有限元格式通过事先构造的近似子空间进行缩减,得到快速计算模型。通过对比贪婪算法和直接正交法在构造近似子空间上的数值计算特性,提出分级自适应子空间构造技术。该计算方法是减基法在结构分析领域的扩展,可以大幅度提高多参数条件下大规模问题反复计算的效率。

关键词:

Abstract:

A hierarchical adaptive technique was suggested to rapidly evaluate the structural behaviors of a multi-parameterized large-scale problem in vehicle body design.It is found that the method is computational stability with a higher accuracy.The stiffness matrix of finite element shell formulation is first decomposed and reformed to achieve a parametrized formulation.The reduced system is obtained by projecting the original system onto a predefined approximate subspace.In order to improve the computational performance in constructing the subspace,the greedy adaptive method and the directly orthogonal method were analyzed and a hierarchical adaptive technique was suggested. It is realized that proposed technique is efficient in structural engineering such as vehicle body design and is applicable to many other structural design contexts.

Key words: vehicle body design, reduced-basis method, adaptive technique, hybrid algorithm

中图分类号: