A step towards ultrasonic guided wave monitoring for resin infusion front position estimation in composites manufacturing

Resin infusion paired with Out of Autoclave (OoA) curing offers an alternative to infrastructure-heavy autoclave-based manufacturing. However, lower fibre volume fractions and increased porosity from uneven resin flow limit the adoption of OoA processes in safety-critical applications. Consequently, there is demand for in-situ monitoring tools to track resin progression and ensure full permeation. Prior methods, including optical fibres and electromagnetic sensors, can infer front position but are intrusive or hard to scale. This research investigates leaky Lamb waves generated by ultrasonic transducers embedded in the top lid of an infusion mould. To isolate wave-fluid interactions, liquid-only measurements in a 2.0 mm thick infusion box are collected, removing laminate heterogeneity and enabling acquisition of controllable consecutive measurements, enabling the development and validation of predictive models under well-defined conditions. Attenuation of the fundamental antisymmetric mode (A0) as resin reaches the sensing region was demonstrated through theoretical and simulation-based analysis, highlighting the potential of Ultrasonic Guided Waves (UGW) for real-time fluid tracking. A custom experimental setup enabled consistent repeatable measurements of an advancing liquid front. A parametric study investigated the effects of geometry and fluid on signal amplitude, determining sensor spacing for sensitivity and areal coverage. Ultrasonic measurements were correlated with time-stamped images of the resin front through a machine-vision algorithm. Several functional approximation methods were applied to estimate liquid position from ultrasonic data, capturing the general trends in flow behaviour. The models yielded robust predictions, with mean errors of 5-7% of the sensor spacing, despite environmental variations and system nonlinearities contributing to data variability. • Confirmation of guided wave energy loss in fluid-coupled thin structures. • Setup enabling liquid-front tracking via network of piezoelectric transducers. • Study defining optimal sensor spacing for efficient areal fluid monitoring. • Real-time machine-vision algorithm tracks and labels liquid front in real-time. • Real-time liquid front estimation from ultrasonic data using a statistical model.