Application of multisynchrosqueezing transform for structural modal parameter identification

The accurate identification of modal parameters is a critical issue in the determination of features of civil structures. In this paper, a novel method based on the multisynchrosqueezing transform (MSST) is proposed to identify modal parameters, including natural frequencies, damping ratios and mode shapes of civil structures. The MSST consists of multiple operations of a synchrosqueezing transform so that the time-frequency representation of an analyzed signal becomes more concentrated, which allows more accurate decomposition of the signal. To identify modal parameters based on the MSST, first, the natural extraction technique is used to obtain a free vibration response from a measured ambient vibration response. Second, the free vibration response is decomposed into several modes by using the MSST, and mode shape vectors can be obtained from the decomposed modes for all measurements. Then, instantaneous phases and instantaneous amplitudes of the modes are obtained by using the Hilbert transform. Finally, a least-squares curve fitting technique is performed on the instantaneous phases and instantaneous amplitudes to extract natural frequencies and damping ratios. Two numerical examples, a 3-degree-of-freedom free vibration response signal and a four-story frame steel structure subjected to environmental vibration, are used to demonstrate the applicability of the MSST-based method. In addition, an experimental validation based on a pedestrian overpass, located in Tufts University, United States, is conducted. Case analyses indicate that the MSST-based method can easily identify high-quality natural frequencies and damping ratios from measurements of structures.