A frequency and spatial domain decomposition method for operational strain modal analysis and its application. (1st May 2016)

• Record Type: Journal Article
• Authors: Wang, T.; Celik, O.; Catbas, F.N.; Zhang, L.M.
• Is Part Of: Engineering structures. Volume 114(2016:May 01)
• Page Start: 104
• Page End: 112
• Publication Date: 2016-05-01
• Subjects: Strain modes -- Operational modal analysis -- Frequency and spatial domain decomposition -- Damage detection; Structural engineering -- Periodicals; Structural analysis (Engineering) -- Periodicals; Construction, Technique de la -- Périodiques; Génie parasismique -- Périodiques; Pression du vent -- Périodiques; Earthquake engineering; Structural engineering; Wind-pressure; Periodicals; 624.105
• DOI: 10.1016/j.engstruct.2016.02.011
• Languages: English
• ISSNs: 0141-0296

Abstract

Highlights: An operational modal analysis algorithm in spatial domain is proposed. Global characteristics of a structure can be obtained by strain modal analysis. Strain gauge based modal analysis are not good at high frequency identification. A minor change of boundary condition was detected by the analysis of strain mode shapes. Abstract: Operational modal analysis (OMA), where only the responses are utilized, has been applied to various engineering problems more commonly due to its advantages in real life implementations. A frequency and spatial domain decomposition method for operational modal analysis making use of strain measurements is presented in this paper. With the proposed algorithm, accurate global characteristics of a structure can be obtained from only the measured strain responses. Singular value decomposition, power spectrum enhancement, and the least square fitting techniques are adopted to decouple and obtain the strain modes one by one. Strain modal analysis and acceleration modal analysis are conducted simultaneously to extract the modal parameters of a four-span bridge model with a pair of heavily coupled modes. The processes and the results are carefully compared with each other. Starting from the abnormal strain modes observed, the accidental change of a boundary condition of the bridge model is successfully detected, and also accurately located by the correlation calculation of strain mode shapes.