Open Collections

Abstract

This thesis investigates the low-velocity impact (LVI) response, post-impact damage morphology, and detection reliability of Prepreg Platelet Molded Composites (PPMC), a class of long discontinuous fibre-reinforced polymers characterized by a stochastic meso-structure. A finite element framework is developed to simulate the LVI response of PPMC plates using voxel-based platelet representations and cohesive interfacial damage models. Statistical validation of simulated force-time and energy-time curves is conducted using functional analysis of variance (fANOVA), demonstrating the model’s ability to capture both average behaviour and variability in experimental data. Post-impact damage is characterized through multi-modal Non-Destructive Testing (NDT), including X-ray Computed Tomography (µCT), Ultrasonic Testing (UT), and Infrared Thermography (IRT). These imaging modalities reveal complex damage patterns driven by inter-platelet delamination, matrix cracking, and fibre failure. A data fusion framework is implemented using both feature integration and classification approaches to enhance visualization and quantification of the damage process zone. To evaluate inspection reliability, a Probability of Detection (POD) analysis is performed using Transient Infrared Thermography on composite plates with artificial Flat-Bottom Hole defects. The results highlight the influence of stochastic thermal behaviour on detectability and demonstrate the need for adapted POD methodologies in discontinuous fibre systems. Overall, this thesis provides a validated numerical-experimental framework for impact analysis and inspection reliability in platelet-moulded composite systems.