Modal parameter identification for closely spaced modes using an Empirical Fourier decomposition-based method

Document Type : Article

Authors

1 Guizhou University

2 Guizhou Construction Science Research & Design Institute Limited Company of CSCEC

3 Guizhou Gongda Science and Technology Service Co. LTD

Abstract

Precise and effective identification of modal parameters is important for a civil structure throughout the structural lifetime. A time-frequency method for modal parameter identification is developed based on a recently developed adaptive method for signal decomposition, i.e., empirical Fourier decomposition (EFD). The EFD is used to separate a multi-component free vibration response into the summation of several mono-component modal responses. The modal frequencies and modal damping ratios are calculated from the modal responses by using the empirical envelope method and a damping estimation method. Two numerical examples and one experimental example are provided to validate the EFD-based time-frequency method and highlight the improvements of the EFD-based method relative to the empirical mode decomposition (EMD)-based method. It is highlighted that the EFD-based method has a much higher frequency resolution and lower computational cost than the EMD-based method. Hence, the proposed EFD-based method is suitable for online modal parameter identifications of structures with closely spaced modes. In addition, the EFD-based method is useful for both linear and nonlinear systems.

Keywords