Structural Modal Parameters Identification Under Ambient Excitation Using Optimized Symplectic Geometry Mode Decomposition

In the process of structural modal parameters identification under environmental excitation, the employed measured dynamic response signal is usually non-stationary and contains noise. As a novel signal analyzed method, symplectic geometry mode decomposition (SGMD) has been proven to be effective for dealing with non-stationary and noisy signals. However, the traditional SGMD may treat noise as modal information, which inevitably undermines the accuracy of modal identification. To overcome this problem, this paper proposes an optimized SGMD algorithm for structural modal parameter identification. First, the SGMD algorithm is refined with the Hankel matrix, Kurtosis theory, Pearson correlation coefficient, and energy entropy theory. Then, the natural frequencies and damping ratios are identified using the proposed method, which consists of the optimized SGMD algorithm, natural excitation technique (NExT), direct interpolating method, and curve-fitting function. Finally, the applicability of the proposed method under environmental excitation is investigated by two examples: a two-story frame structure and an 88-story office tower. The results demonstrate that the proposed method is effective for identifying the civil structural modal parameters from non-stationary dynamic responses.