Experimental Study and Artificial Intelligence-Based Modeling of the Indentation Behavior of Adhesive-Bonded Curved Composites

Abstract

The aim of this research is to investigate the effect of patch layup stacking, thickness, and radius of curvature on the energy absorption and load-carrying capacity of adhesive-bonded curved glass fiber laminated composite specimens under indentation loads. A testing machine with a hemispherical indentation apparatus was used to evaluate the damage resistance of curved composites. The indentation resistance increased with increasing radius of curvature in both monolithic and bonded curved composite specimens. The specimens with [0/90]n orientation exhibited higher absorbed energy values and indentation resistance compared to those with [+/- 45]n orientation. The number or thickness of fibers in the upper and lower layers used in the repair process influenced the energy absorption capacity. Applying thicker layers in the indentation direction of bonded curved composite specimens produced superior indentation resistance. The degree of delamination damage was influenced by the curvature diameter and layer configuration, with delamination dominant in bonded samples. Additionally, an artificial neural network was used to predict the indentation responses of bonded curved composite specimens. The ANN model accurately simulated the force-displacement curves and peak forces of monolithic and bonded curved laminates.