| 动力特征向量灵敏度的组合扰动迭代算法 | |
| 曾国华 ; 董聪 ; Zeng Guohua ; Dong Cong | |
| 2010-06-07 ; 2010-06-07 | |
| 关键词 | 特征向量 灵敏度 迭代方法 重特征值 eigenvector sensitivity iterative method repeated eigenvalues O302 |
| 其他题名 | Iterative Algorithm with Combination and Perturbation Behavior on Sensitivity Analysis of Dynamic Eigenvector |
| 中文摘要 | 为了克服重动力特征值系统中,求解特征向量灵敏度的传统算法需要Hessian矩阵信息、计算步骤复杂等各种不足,通过组合系数将基于动力特征方程和质量矩阵正则化条件推导所得的特征向量灵敏度约束方程统一成满秩的迭代形式,并引入基于扰动系数和质量矩阵的扰动矩阵,改善统一迭代形式的收敛性能。理论分析和实例研究表明,该方法便于程序实现,同时只要选择合适的组合和扰动系数,仅需迭代几步就能收敛到特征向量灵敏度的满意解,且收敛性能与初始点的选择无关。; To overcome the drawbacks of traditional algorithms for dynamic eigenvector derivatives with repeated eigenvalues,such as complexity procedure and need for Hessian matrix,the constraints of eigenvector derivatives derived from the equations of dynamic characteristics and the normalization with mass matrix are combined into a uniform iterative form.A perturbed matrix based on perturbed coefficients and mass matrices is also incorporated into the iterative form to improve its convergence property.Under the theoretical and numerical analysis,it is concluded that the proposed algorithm is suitable for programming,and if the coefficients of combination and perturbation are selected properly,it can give satisfactory results after one or a few iterations.Its convergence property is independent of the initial points.; 国家自然科学基金资助项目(编号:50278047); 国家“八六三”基金资助项目(编号:2002AA615110); 北京市自然科学基金重点资助项目(编号:8021002); 北京市重大科技基金资助项目(编号:H030630210021) |
| 语种 | 中文 ; 中文 |
| 内容类型 | 期刊论文 |
| 源URL | [http://hdl.handle.net/123456789/46502]  |
| 专题 | 清华大学 |
| 推荐引用方式 GB/T 7714 | 曾国华,董聪,Zeng Guohua,等. 动力特征向量灵敏度的组合扰动迭代算法[J],2010, 2010. |
| APA | 曾国华,董聪,Zeng Guohua,&Dong Cong.(2010).动力特征向量灵敏度的组合扰动迭代算法.. |
| MLA | 曾国华,et al."动力特征向量灵敏度的组合扰动迭代算法".(2010). |
| 个性服务 |
| 查看访问统计 |
| 相关权益政策 |
| 暂无数据 |
| 收藏/分享 |
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。
修改评论