Analysis of electrical degradation of aerospace carbon fibre composites using ultrasonic testing and machine learning

Aircraft developments have increased reliance on Carbon-Fibre Reinforced Plastic (CFRP) composites and more electrical systems, making the electrical behaviour of CFRPs a crucial factor in ensuring safe performance. Machine Learning (ML) algorithms are gaining unprecedented traction and adoption for Non-Destructive Evaluation (NDE) data analysis of composite systems. This research develops a diagnostic framework that integrates machine learning with ultrasonic NDE to assess effects of electrical loading conditions on CFRP for the early detection and characterization of electrical degradation. Initial ultrasonics testing and ML results indicate threshold temperatures and associated current levels for delamination.