论文部分内容阅读
为了提高随机子空间法的识别速度,采用MAC准则优选数据精简Hankel矩阵的“过去”输出数据和通过简化模态参数识别步骤的方法,推导了随机子空间优化算法,并借助Matlab平台编写程序以达到快速化识别的目的。其一精简Hankel矩阵“过去”输出数据的同时有效地避免了模态遗漏;其二详尽分析了Hankel矩阵QR分解得到的子矩阵R_(21),将可观测矩阵与矩阵R_(21)的奇异值分解建立直接关系,避免求解投影矩阵。研究结果表明:使用部分数据作为“过去”输出数据,减少了计算量;避开求解投影矩阵,简化了计算步骤;避免高维矩阵的存储和分解,很大程度上改善了计算机的使用内存;识别速度增幅明显,精度与其他文献相吻合。最后以西宁北川河桥为工程算例,验证了该优化算法的实用性和有效性,得到比较理想的结果。
In order to improve the recognition speed of stochastic subspace method, the “past” output data of Hankel matrix is reduced by using MAC criterion optimization data and the stochastic subspace optimization algorithm is derived by simplifying the method of modal parameter identification. Program to achieve the purpose of rapid identification. In the first part, the modal omission is effectively avoided while the output data of Hankel matrix “past ” is reduced. The second part analyzes the submatrix R_ (21) obtained from the QR decomposition of Hankel matrix in detail and combines the observable matrix with the matrix R_ (21) The establishment of a direct relationship between the singular value decomposition, to avoid solving the projection matrix. The results show that the use of partial data as output data of “past ” reduces the amount of computation, avoids solving the projection matrix, simplifies the calculation steps, avoids the storage and decomposition of high-dimensional matrix and greatly improves the use of the computer Memory; identification speed increased significantly, accuracy and other literature. Finally, Xining Beichuan River Bridge is taken as an example to verify the practicability and effectiveness of this optimization algorithm, and get the ideal result.