To reduce fuel consumption of modern aircrafts, polymeric composite materials are considered for structural components such as wings, cockpit, stabilizer and fuselage. The performance of composite materials can deteriorate during the aircraft service life as a results of matrix cracking, fiber failure and puncture or delamination. Efficient detection, quantification and localization of damage as well as accurate prediction of the deterioration is central to making polymeric composites a safe and cost-effective material. One strategy to achieve this can be built-in Structural Heath Monitoring (SHM) systems that analyze structural response features to detect damages.
Such SHM strategies are only feasible if they can accurately detect damages (probability of detection POD) while limiting the number of false alarms (false alarm rate FAR). The aim of this project was therefore to utilize state-of-the-art probabilistic modelling tools, machine learning techniques and decision analysis to investigate the required accuracy of SHM and to develop algorithms and software for improved predictions based on laboratory tests.
Models for determining the effect of monitoring on the reliability of aircraft wing panels.
Optimization of replacement strategy and inspection for aircraft structures based on structural reliability concepts.
Quantification of the economic effect of SHM on the asset integrity management costs of an aircraft.
Software for structural damage detection in composite panels with built-in sensors network through multilevel decision fusion.