Baseline-free structural damage identification for plate-like structures based on two-dimensional curvature propagating flexural waves

Abstract

Waveforms of propagating flexural waves can reveal plentiful information about local anomalies caused by damage, and the local anomalies can be used for damage identification. However, the local anomalies can be masked by the interference of measurement noise and global trends of the waveforms, and they may be able to indicate only fractions of the size and extent of the damage. In this paper, an effective noise-robust damage identification method for plate-like structures is proposed based on the fact that damage can introduce imminent local anomalies to two-dimensional curvature propagating flexural waves (2D-CPFW). The 2D-CPFW can be alternated using two-dimensional continuous wavelet transform (2D-CWT) to lower adverse effects of measurement noise and errors. To suppress global trends and intensify the local anomalies in the 2D-CPFW, 2D-CWT with a higher-order Laplacian of Gaussian function and the Teager energy operator are applied. The fundamental mechanism of how 2D-CWT with different orders of Laplacians of Gaussian function can suppress global trends of 2D-CPFW is investigated for the first time. It is found that the rth-order Laplacian of Gaussian function can well suppress the global trend of a 2D-CPFW within a finite interval if the trend can be well approximated by a 2r−1th-order bi-variant polynomial. The modal assurance criterion and a statistical criterion are used to assist the selection of the proper order of Laplacian of Gaussian function. An accumulative damage index is proposed, and the locations and extent of the damage can be identified within neighborhoods with high damage index values. If an intact plate-like structure is geometrically smooth and made of materials that have no mass and stiffness discontinuities, the proposed method does not require any waveforms of the intact structure serving as a baseline, and therefore, the method can be considered baseline-free. Effectiveness and noise-robustness of the proposed method are investigated in two numerical examples. Two experimental investigations were conducted on two aluminum plates under different damage scenarios. Both the numerical and experimental examples verify that the proposed method is effective and noise-robust in identifying the location and extent of the damage.