Experimental tensile testing of the lap joint composite laminates supported with the acoustic emission and machine learning techniques

This paper investigates tensile behavior of through thickness reinforced carbon/epoxy lap joint composite laminates, reinforced with steel z-pins and staples, arranged in two rows parallel to the overlapping edges, via experimental testing. Acoustic emission (AE) monitoring is employed during the displacement-controlled tensile tests to monitor damage propagation during loading using the Vallen AMSY-5 measurement system, with two piezoelectric sensors being mounted at the laminate surface. Furthermore, machine learning algorithms are integrated to process AE data, enabling the recognition and prediction of failure mechanisms. Fractographic analyses were performed to observe the nature of damage post-failure. The experimental research was enriched with capturing high-resolution pictures of total crack propagation length growth using a high-resolution photocamera. The performed empirical tests demonstrated that the unstable propagation of a crack along the bonding interface has led to an eventual breakdown of both unreinforced and reinforced joints. An increase in the full displacement and load at failure was clearly detected for both z-pins and staples with a noticeable decrease in crack growth length, while a higher performance was illustrated for staples in comparison to z-pinned and unpinned.