Damage location, quantification and characterization of steel-concrete composite beams using acoustic emission

In the hogging moment regions of steel-concrete composite beams that have been widely used in bridges and buildings nowadays, the concrete slabs are vulnerable to cracks and the connection interfaces are subject to debonding. These damages could substantially reduce the stiffness, strength and durability of the beams. Comprehensive and quantitative structural health monitoring of steel-concrete composite beams is in great de-mand. To this end, a novel framework was developed in this study for damage location, quantification and characterization of steel-concrete composite beams using acoustic emission (AE) measurements. The damage -induced AE sources were located more accurately based on empirical wavelet transform (EWT), Akaike infor-mation criterion (AIC) and genetic algorithm (GA), where EWT and AIC together helped to determine the exact arrival times of AE waves in the presence of serious attenuation, reflection, scattering and dispersion of AE waves and high operational noise of the structure. With the coordinates of located AE events, the macrocrack and debonding surfaces were estimated by hybrid hierarchical-k-means clustering analysis. After that, the micro -cracking modes and orientations were diagnosed based on moment tensor analysis that made use of the co-ordinates of AE events and the P-wave first motions. Through the reversed four-point bending test of steel -concrete composite beams, the proposed framework was proved to be able to detect both concrete cracks and interface debondings. Multiple damages of different types could be accurately located and quantified. The dominant damage mechanism of cracks in the concrete slabs was found to be tensile-mode microcracking, and the interface debonding was also primarily caused by tensile stress-related vertical uplift.

Automated summary

A novel framework was developed in this study for damage location, quantification and characterization of steel-concrete composite beams using acoustic emission (AE) measurements. The damage-induced AE sources were located more accurately based on empirical wavelet transform (EWT), Akaike information criterion (AIC) and genetic algorithm (GA). With the coordinates of located AE events, the macrocrack and debonding surfaces were estimated by hybrid hierarchical-k-means clustering analysis.