Baseline-free bridge damage identification using proper orthogonal decomposition and continuous wavelet transform with limited sensors

Abstract

Traditional modal identification methods usually fail to provide high-resolution mode shape estimations due to insufficient sensor observations, thus significantly reducing their effectiveness. This study introduces a novel baseline-free bridge damage identification method that requires a minimal number of sensors. An analysis method, named POD-CWT transform method, is introduced. This method incorporates proper orthogonal decomposition (POD) and continuous wavelet transform (CWT). It can effectively identify the locations of damage in girder bridges subjected to loads from moving vehicles. The displacement transducers are arranged in an equidistant manner and used to measure the bridge displacement responses under moving loads. POD is employed to obtain a component matrix, wherein each column represents the vibration mode and dynamic information of the bridge. High-resolution mode shapes are obtained by filtering this dynamic information through a low-pass filter. The first mode is processed using CWT to extract the wavelet energy content, which serves as the structural damage indicator. When structural damage is present, the corresponding wavelet energy content profile at the damaged location undergoes an abrupt change. The effectiveness of this approach is verified by numerical simulations and experimental tests. The presented algorithm eliminates the need for non-damage baseline data and significantly reduces the number of required sensors without sacrificing estimation accuracies.