Quantitative acoustic emission investigation on the crack evolution in concrete prisms by frequency analysis based on wavelet packet transform

Abstract

To gain an insight into the evolution of micro-cracks in concrete materials, a quantitative acoustic emission investigation on the damage process of concrete prisms subjected to three-point bending loading was performed. Each of the monitored acoustic emission signals was processed by a two-level wavelet packet decomposition into four different frequency bands (AA2, DA2, AD2, and DD2), and the energy coefficients R1, R2, R3, and R4 that parameterize their characteristic frequency bands were calculated. By analyzing variations in energy coefficients of the lowest frequency band (AA2), R1, and the energy coefficients of the highest frequency band (DD2), R4, the whole damage process was divided into three stages: crack initiation, crack growth, and crack coalescence. An inverse relationship between the frequency of the acoustic emission signal emitted by the propagating crack and the crack size in concrete materials was acquired based on the damage theory of brittle materials and the strain energy release theory. The statistical analysis results of the experimental data indicated that the average of R1 increased in turn, and the average of R4 correspondingly decreased in turn from Stage 1 to Stage 3. It revealed that the frequencies of acoustic emission signals decreased gradually with the evolution of the damage of concrete prisms, which is in a good agreement with the theoretical analysis result.

Looseness monitoring of multiple M1 bolt joints using multivariate intrinsic multiscale entropy analysis and Lorentz signal-enhanced piezoelectric active sensing

“…To compare our proposed approach with other strategies using different feature extraction algorithms and machine learning methods, comparison studies are further conducted. Wavelet packet decomposition (WPD) 60 and complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) 49 are also utilized to decompose the three-dimensional signal, namely x, y , and z components of the collected Lorentz signal, separately. The spectral density (SD), along with MSE and IMSE are adopted as condition indicators.…”