Damage mode identification of adhesive composite joints under hygrothermal environment using acoustic emission and machine learning

Abstract This paper studies the hygrothermal aging effect on the damage behaviors of adhesive composite joints by acoustic emission (AE) technique. Tensile tests and AE tests are launched for both non-aging and aging specimens under regular and hygrothermal environments, respectively. The clustering analysis, the time-domain analysis and the frequency-domain analysis are combined to identify various damage modes and to study the hygrothermal aging effect of single-lap joint (SLJ). First, the clustering analysis is performed to study the damage mode correlation using four AE characteristic parameters including the time of duration, the peak amplitude, the RA value (the rise time divided by the peak amplitude) and the frequency centroid of gravity. Then, the time-domain analysis is conducted to investigate the hygrothermal effect of the single-lap joint (SLJ). Finally, the frequency spectrum analysis is carried out to study the frequency range for the shear failure of adhesive layer. In addition, the dominant damage mode of SLJ is identified using the wavelet-based decomposition of the ultimate AE signals with the largest energy.