线性特征值问题的高精度矩阵摄动法

High-Accuracy Matrix Perturbation Method of Linear Eigenvalue Problems

线性特征值问题摄动重分析的基本理论已趋于成熟。本文进一步探讨了提高摄动计算精度的策略,提出了一种分步摄动一摄动迭代算法。通过实例计算表明,该算法可在结构具有大修改量的情况下有效地实现广义特征值问题的高精度重分析。

The perturbation reanalysis theory of linear eigenvalue problem Ax=λBx isbeing perfected.The authors further study in this paper the computationtechniques for enhancing the accuracy of perturbation estimate solutions,and astep-by-step perturbation-perturbational iteration algorithm is presented.Severalnumerical examples show that the algorithm is very effective to hlgh-accuracyreanalysis of the generalized eigenvalue problems with large changes in parametermatrices A and B.