Extreme-point Symmetric Mode Decomposition-based Energy Integral Model for Bridge Abnormality Detection Using Ground-based Synthetic Aperture Radar

Ground-based synthetic aperture radar (GB-SAR), as a noncontact measurement technology, is widely used in the dynamic deflection monitoring of various bridges. Energy analysis is a popularized time-frequency domain technique for bridge abnormality detection. To improve the accuracy of bridge abnormality detection using GB-SAR, in this paper, we propose an extreme-point symmetric mode decomposition (ESMD)-based energy integral model based on the total energy function to identify the position and trend of bridge changes. First, ESMD with the wavelet synchro-squeezing transform (ESMD-WSST) high-frequency denoising processes are applied to reduce the effect of noise contained in the monitored dynamic deflection. Second, ESMD decomposition and instantaneous frequency calculation are performed on the denoised signal to obtain the integration time adaptively. Third, the instantaneous total energy of all reflection points on the lower surface of the bridge is calculated through the kinetic energy formula to improve the accuracy. Finally, instantaneous total energy integration is applied for energy accumulation calculation to accurately detect bridge abnormality without empirical judgment. The performance of the proposed model is verified through an on-site experiment of the Beishatan Bridge and comparing its result with that of the 3D laser model in the same period. The experimental results show that the proposed model can achieve high-precision identification of bridge abnormality positions and trends.

Automated summary

In this paper, an energy integral model based on extreme-point symmetric mode decomposition (ESMD) is proposed to improve the accuracy of bridge abnormality detection using GB-SAR. The model utilizes denoising processes, ESMD decomposition, and instantaneous frequency calculation to enhance the precision of bridge abnormality identification without empirical judgment.