Modal Parameter Identification Based on an Enhanced Hilbert Vibration Decomposition

Abstract

Hilbert vibration decomposition (HVD) is an effective tool for decomposing multi-component signals into several mono-component signals, while HVD may result in a poor decomposition when the energies of two or more modes are close. In this paper, a masking signal enhanced HVD is proposed and further utilized for modal parameters identifications of civil and mechanical systems. The enhanced HVD is used to decompose structural vibrational signals into mono-component modal responses. The empirical envelope method is then used to calculate the instantaneous frequencies and instantaneous damping ratios of the modal responses. Several numerical and experimental examples are analyzed to demonstrate the superiority of the enhanced HVD relative to the original HVD and the well-known empirical mode decomposition, by which the accuracy of the modal parameter identification method is also validated. The enhanced HVD is proved to outperform the original HVD and the empirical mode decomposition in decomposing certain types of signals. The present modal parameter identification technique is efficient and accurate for typical linear and nonlinear structures, including systems with closely spaced modes, low energy modes, amplitude-dependent modal parameters, etc. Therefore, the present modal parameter identification technique can be a useful tool for the identification of large-scale civil structures, e.g., long-span bridges.