An Improved Structural Health Monitoring Method Utilizing Sparse Representation for Acoustic Emission Signals in Rails

The structural health of rails significantly impacts the safety of railway transport, which requires long-term and accurate monitoring. An improved structural health monitoring (SHM) method is proposed to enhance the damage information and construct an adaptive weighted index for evaluating the structural health states of rails. In this method, modified singular value decomposition (SVD) with a double-layer Hankel matrix compensation algorithm enhances the damage information and ensures real-time detection. Based on the inflection point of the inner product gradient, the step size of the sparsity adaptive matching pursuit (SAMP) algorithm is adaptively adjusted to reconstruct damage signals. Subsequently, an innovative index is constructed with adaptive weighting for SHM of rails. Tensile tests are performed to verify the effectiveness of the proposed method. The results show that the proposed method avoids the defects of traditional nondestructive testing (NDT) techniques that rely on a priori information about the damage signal (such as damage frequency acoustic emission (AE) hits rate and waveform characteristics). Furthermore, using the variation in signal energy, the proposed method has the capability to accurately delineate the three structural health stages under complex waveform conditions, which is not possible with conventional detection methods. This method provides guidance for the damage analysis and monitoring techniques of rails.