An improved EMD method for modal identification and a combined static-dynamic method for damage detection
An improved EMD method for modal identification and a combined static-dynamic method for damage detection
复制标题
改进的 EMD 模态识别方法和静动态结合损伤检测方法
DOI：
10.1016/j.jsv.2018.01.036
复制
发表时间：
2018-04
影响因子：
4.7
通讯作者：
Xu Dian
中科院分区：
文献类型：
--
作者：
Yang Jinping;Li Peizhen;Yang Youfa;Xu Dian
Empirical mode decomposition (EMD) is a highly adaptable signal processing method. However, the EMD approach has certain drawbacks, including distortions from end effects and mode mixing. In the present study, these two problems are addressed using an end extension method based on the support vector regression machine (SVRM) and a modal decomposition method based on the characteristics of the Hilbert transform. The algorithm includes two steps: using the SVRM, the time series data are extended at both endpoints to reduce the end effects, and then, a modified EMD method using the characteristics of the Hilbert transform is performed on the resulting signal to reduce mode mixing. A new combined static-dynamic method for identifying structural damage is presented. This method combines the static and dynamic information in an equilibrium equation that can be solved using the Moore-Penrose generalized matrix inverse. The combination method uses the differences in displacements of the structure with and without damage and variations in the modal force vector. Tests on a four-story, steel-frame structure were conducted to obtain static and dynamic responses of the structure. The modal parameters are identified using data from the dynamic tests and improved EMD method. The new method is shown to be more accurate and effective than the traditional EMD method. Through tests with a shear-type test frame, the higher performance of the proposed static-dynamic damage detection approach, which can detect both single and multiple damage locations and the degree of the damage, is demonstrated. For structures with multiple damage, the combined approach is more effective than either the static or dynamic method. The proposed EMD method and static-dynamic damage detection method offer improved modal identification and damage detection, respectively, in structures.
登录
查看更多内容
影响因子：
2.6
作者：
W. Yan;Tian-Li Huang;W. Ren
通讯作者：
W. Yan;Tian-Li Huang;W. Ren
影响因子：
5.4
作者：
E. Sevillano;R. Sun;R. Perera
通讯作者：
E. Sevillano;R. Sun;R. Perera
影响因子：
5.5
作者：
L. Mao;Yong Lu
通讯作者：
L. Mao;Yong Lu
DOI：
10.1061/(asce)0733-9399(2008)134:11(1000
发表时间：
2008-11
期刊：
Journal of Engineering Mechanics-asce
影响因子：
--
作者：
T. Hsu;C. Loh
通讯作者：
T. Hsu;C. Loh
DOI：
10.1016/j.dsp.2016.07.012
发表时间：
2016-11
期刊：
Digit. Signal Process.
影响因子：
--
作者：
Rajib Sharma;S. Prasanna
通讯作者：
Rajib Sharma;S. Prasanna