Data Visualization Approach for Operational Loading Variations of an Aircraft Wing Box using Vibration based Damage Detection
Abstract »Structural health monitoring (SHM) is gaining recognition among aerospace market players for structural performance during flight testing as well as for condition based maintenance (CBM) implemented while aircrafts are in service . However, one of the key challenges in developing a robust damage detection system for aircrafts in service is the capability to discriminate the effects of operational and environmental variations from damage-sensitive features. This paper explores an innovative experimental method in establishing operational loading variability inspired by varying aircraft fuel loading conditions by adopting FRF vibration data to develop a pattern recognition. A pattern recognition technique through machine learning algorithm is applied to assist in detecting damage in a simulated aircraft wing box under variable operational loading conditions using a supervised learning technique. Through a data visualization approach based on maximising variance, it is shown that principal component analysis (PCA) effectively results in a consistent identification of the damage classes throughout all operating loading conditions. The promising results of this study illustrates a robust damage detection method encompassing level 1 (detecting damage) and level 3 (diagnosing damage severity) under the influence of operational loading variations by adopting linear PCA and multivariate statistical analysis.
Biography: A PhD student and a research group member in Dynamics Research group at the University of Sheffield. The author is focusing on vibration based method under the effects of operational loading variability for aircraft structures using statistical pattern recognition based and machine learning algorithm techniques.
Affiliation: University of Sheffield (UOS) Mechanical Engineering S1 3JD Sheffield United KingdomManson, GraemeManson, Graeme firstname.lastname@example.org
Affiliation: University of Sheffield Mechanical Engineering Sheffield United KingdomWorden, KeithWorden, Keith email@example.com
Affiliation: University of Sheffield Mechanical Engineering S1 3JD Sheffield United Kingdom