Research on structural damage identification methods for aircraft full scale fatigue test

icaf2023 Tracking Number 80

icaf2023 poster

In aircraft full scale fatigue test, identify and locate structural fatigue cracks accurately is an important job, which traditionally was completed by regular Non Destructive Inspection (NDI). While, due to the high complexity of structure geometry and the limitation of detection space, fatigue cracks found by NDI tend to be larger in size or even missed in detection, which is extremely disadvantageous for structural repair. In order to solve the above problems, several methods such as threshold method, relative error method, probability distribution method and slope method have been presented. However, it is found that these methods have insufficient ability to identify short cracks and poor adaptability to strain data of different masses in the application of several full scale fatigue tests. In view of the above problems, a strain sensor based method on fatigue damage monitoring and identification for critical locations of aircraft structure in full scale fatigue test was established, on the foundation of high accuracy stress modelling analysis. Firstly, an approach used to distinguish failed strain sensors was presented in order to avoid false damage warning. Secondly, a data cleaning method which combined K-means clustering algorithm with 3σ rule was developed and used to detect and eliminate outliers in sensor data. Lastly, a fatigue damage identification method with adaptive threshold was proposed. In aircraft full scale fatigue test, several fatigue cracks on critical structures were identified and located timely and successfully using these methods. More importantly, it is more accurate and sensitive compared with existing methods, such as threshold method, probability distribution method, relative error method and slope method. Methods presented in this paper were significant for structural design modification and crack repair cost reduction. It can also be applied to structural health monitoring.