Identification of operational modal parameters for a large floodgate under hydraulic excitation using the blind source separation improved by Hankel matrix singular value and adaptive variational mode decomposition

The hydraulic excitation for the discharge structure under operational conditions is difficult to measure directly and accurately. Therefore, obtaining the dynamic characteristics of the structure in complex environments and under conditions of insufficient a priori information is vital to ensure the long-term safe operation of the discharge structure. However, the identification results of the discharge structure's modal parameters are susceptible to noise interference, and the lack of a priori information affects the accuracy of the identification results. Therefore, these problems are addressed by proposing an improved blind source separation based on Hankel matrix singular value and adaptive variational mode decomposition (HMSV-AVMD) method for operational modal parameter identification of large floodgates under hydraulic excitation. First, the HMSV-AVMD method is used to denoise the time-history response signals of multiple measurement points in the structure. Then blind source separation is performed on the denoised signals by the second-order blind identification algorithm. The separation results can determine the order of the structural system and identify modal parameters. The simulated signals and the numerical test results of the system show that the proposed method can effectively achieve adaptive filtering of the background white noise and low-frequency noise in the signals. Moreover, the modal parameter identification results are closer to the theoretical values than the results of other methods. The relative errors of the frequency identification results and the damping ratio identification results are within 2% and 15%, respectively. The proposed method is applied to the modal parameter identification of a large floodgate. Results show that, unlike other methods, the proposed method can achieve system order determination and accurate identification of dense frequency modes for large floodgates.