基于稀疏矩阵技术的多体动力学仿真快速算法.pdfVIP

第 21 卷第 14 期 系 统 仿 真 学 报© Vol. 21 No. 14 2009 年 7 月 Journal of System Simulation Jul., 2009 基于稀疏矩阵技术的多体动力学仿真快速算法 高海涛，张志胜，曹 杰，史金飞 （东南大学机械工程学院，南京 211189 ） 摘 要：在分析多体动力学仿真计算中广泛使用的增广法基础上，提出了一种基于稀疏矩阵技术 的改进算法。算法首先修改了拉格朗日乘子的求解方程，并利用系数矩阵的稀疏性进行排序和符号 分解，获得最佳计算数据结构，从而快速求解拉格朗日乘子；针对违约现象，建立直接违约修正方 程，利用上述数据结构快速求解违约修正值。最后,通过一个算例验证了该算法的有效性，并与传 统增广法进行比较，显示了该算法快速和高精度的特点。 关键词：增广法；稀疏矩阵；动力学仿真；算法 中图分类号：TP391 文献标识码：A 文章编号：1004-731X (2009) 14-4238-05 Fast Algorithm of Multibody Dynamic Simulation Based on Sparse Matrix Technique GAO Hai-tao, ZHANG Zhi-sheng, CAO Jie, SHI Jin-fei (College of Mechanical Engineering, Southeast University, Nanjing 211189, China) Abstract: On the basis of analysis on augmentation approach widely used in the multibody dynamic simulation, an improved algorithm based on sparse matrix technique was proposed. At first, solving Lagrange multiplier equations were modified. Sort and symbol decomposition of coefficient matrix based on sparsity was done to obtain a best data structure, then Lagrange multiplier was quickly solved. Aiming at the violation phenomenon, direct equations of violation correction were built, and direct modification was done by applying this data structure to solve correction values again. Finally, compared with traditional augmentation approach through a numerical example, the result shows the algorithm is of high precision and fast speed. Key words: augmentation approach; sparse matrix; dynamic simulation; algorithm 引 言1