Modal Parameter Identification of Bridge Structures Based on EEMD-Wavelet Threshold Denoising

To investigate the vibration response characteristics of long-span cable-stayed bridge structures under environmental excitation, pickup sensors are used to monitor the dynamic deformation of the structures. Given the presence of noise in the monitoring signal, the filtering method of ensemble empirical mode decomposition (EEMD)combined with wavelet threshold denoising is proposed to improve the signal accuracy. First, EEMD is used to decompose the monitoring signal. On the basis of the double criteria of the mean periodic diagram and correlation coefficient methods, the false components are removed and the remaining IMF components are reconstructed. Then, the wavelet threshold denoising method is used to conduct secondary noise reduction of the reconstructed signal. Finally, the RDT-ITD method is used to identify the structural modal parameters. This combined method is applied to process the vibration response of Tianjin Yonghe Bridge, and the finite element model of the structure is built for comparison. The field measurement and finite element simulation results show that the combined filtering method of EEMD and wavelet threshold denoising is better than a single method and can further improve the signal accuracy. The first three vertical natural frequencies and the corresponding damping ratios of the structure are extracted successfully. The natural frequency results obtained from field measurement are consistent with the finite element simulation results. Moreover, the difference of the natural frequency for the first order is 3.07%, which verifies the validity of the combined method. © 2020, Editorial Board of Journal of Tianjin University(Science and Technology). All right reserved.