点击次数：

论文类型：期刊论文

发表时间：2017-09-25

发表刊物：STRUCTURAL ENGINEERING AND MECHANICS

收录刊物：SCIE、EI、Scopus

卷号：63

期号：6

页面范围：743-750

ISSN号：1225-4568

关键字：eigensystem realization algorithm (ERA); spurious mode; stabilization diagram; Hankel matrix; singular value decomposition (SVD)

摘要：Modal parameter identification plays a key role in the structural health monitoring (SHM) for civil engineering. Eigensystem realization algorithm (ERA) is one of the most popular identification methods. However, the complex environment around civil structures can introduce the noises into the measurement from SHM system. The spurious modes would be generated due to the noises during ERA process, which are usually ignored and be recognized as physical modes. This paper proposes an improved stabilization diagram method in ERA to distinguish the spurious modes. First, it is proved that the ERA can be performed by any two Hankel matrices with one time step shift. The effect of noises on the eigenvalues of structure is illustrated when the choice of two Hankel matrices with one time step shift is different. Then, a moving data diagram is proposed to combine the traditional stabilization diagram to form the improved stabilization diagram method. The moving data diagram shows the mode variation along the different choice of Hankel matrices, which indicates whether the mode is spurious or not. The traditional stabilization diagram helps to determine the concerned truncated order before moving data diagram is implemented. Finally, the proposed method is proved through a numerical example. The results show that the proposed method can distinguish the spurious modes.