Development of NDT Knowledge Systems to Improve the Robustness of Inspection of Composite Products

Abstract

Non-Destructive Testing (NDT) is a quality control measure designed to ensure the safety of products according to established variability thresholds prescribed by design requirements. With the development of advanced technologies, The National Composite Centre, Bristol, has identified that the increasing complexity of composite products will lead to severe inspection challenges. Technical limitations and a lack of formalised understanding of the state-of-the-art of NDT of composites will only exacerbate the bottleneck in engineering operations if not addressed.
To address this apparent knowledge gap and understand NDT system complexity, this thesis presents a formulaic approach to introduce intelligence and improve robustness of NDT operations. This is achieved through the systemic development of a high-fidelity Knowledge Base (KB), based upon a modified Lean Six Sigma framework: Define-Measure-Analyse-Improve-Verify. For state-of-the-art technology mapping, a capability matrix that maps material, component, and defect configuration to capabilities and limitations of selected detection methods is established. Population and data validation is demonstrated through experimental testing of controlled reference standards and evaluated against an assessment criteria. System complexity is investigated in ultrasonic testing operations, focussing on capturing inherent risks in the detection of defects and the designation of evidence-based plans for automation platforms.
The KB presents a formalised framework for the documentation of capabilities and limitations of detection methods with respect to component configuration; when coupled with industry requirements, the KB can assist in road-mapping the development of techniques through resource prioritisation. Exploiting the KB in inspection operations will introduce system intelligence, where captured validated applicability data supports knowledge-based decisions for optimising inspection plans. Additionally, the KB highlights the need for Design for Inspection, providing measurable data that should not be ignored. When employed in industry, the methodology will seek to drive improved intelligence in inspection operations, capability and productivity gains, and support the transition towards NDT 4.0.

Date of Award	24 Jan 2023
Original language	English
Awarding Institution	University of Bristol
Sponsors	National Composites Centre
Supervisor	Carwyn Ward (Supervisor), Anthony J Croxford (Supervisor) & Rob Rose (Supervisor)