邵松标, 刚宪约, 李双, 柴山. 一种求解弹性接触问题的缩减二次规划方法[J]. 力学与实践, 2014, 36(5): 604-610. DOI: 10.6052/1000-0879-13-537
引用本文: 邵松标, 刚宪约, 李双, 柴山. 一种求解弹性接触问题的缩减二次规划方法[J]. 力学与实践, 2014, 36(5): 604-610. DOI: 10.6052/1000-0879-13-537
SHAO Songbiao, GANG Xianyue, LI Shuang, CHAI Shan. A REDUCED QUADRATIC PROGRAMMING METHOD FOR ELASTIC CONTACT PROBLEMS[J]. MECHANICS IN ENGINEERING, 2014, 36(5): 604-610. DOI: 10.6052/1000-0879-13-537
Citation: SHAO Songbiao, GANG Xianyue, LI Shuang, CHAI Shan. A REDUCED QUADRATIC PROGRAMMING METHOD FOR ELASTIC CONTACT PROBLEMS[J]. MECHANICS IN ENGINEERING, 2014, 36(5): 604-610. DOI: 10.6052/1000-0879-13-537

一种求解弹性接触问题的缩减二次规划方法

A REDUCED QUADRATIC PROGRAMMING METHOD FOR ELASTIC CONTACT PROBLEMS

  • 摘要: 基于势能原理以节点位移为设计变量、以接触条件为约束方程构建了无摩擦弹性接触问题的二次规划数学模型,在此基础之上利用力平衡线性约束方程的特解和由基础解向量构成的奇异模态矩阵,提出一种新的基于奇异坐标变换的自由度缩减方法,大大降低了二次规划的规模,并使得二次规划模型不再含显性等式约束;根据弹性接触力学体系的特点,通过人为假定接触自由度位移模式,提出了一种简单高效的奇异模态矩阵的计算方法。通过两圆柱接触、轴孔间隙配合接触两个数值算例的对比分析,验证了对于弹性接触问题的求解,缩减二次规划方法有效克服了传统方法计算量大、对求解参数设置敏感、收敛困难的问题。

     

    Abstract: Based on the potential energy principle, a frictionless elastic contact problem quadratic program-ming model is constructed by treating the nodal displacements as the variables and the contact conditions as the constraint equation. On the basis of the model, the particular solution and the base solutions of the linear constraint equation of the force balance are used to construct the singular modal matrix, and then the freedom reduction method based on a singular coordinate transformation is proposed, which greatly reduces the dimension of the quadratic programming, and leaves the quadratic programming with no dominance equality constraint. According to the characteristics of the elastic contact mechanical system, and the contact freedom displacement mode generally assumed,a simple and eflcient singular modal matrix calculation method is pro-posed. The solution of the elastic contact problem is verified through comparison and analysis of two numerical cases of the contact problems of two cylinders and the axle hole clearance fit. The reduction of the quadratic programming method overcomes the problems of large amount of work, the sensitivity to the parameter settings and the convergence diflculties inherent in the traditional method.

     

/

返回文章
返回