An Enhanced Guided Wave-Gaussian Mixture Model for Aircraft Structural Damage Monitoring under Varying Environmental and Operational Conditions

Abstract

During the past two decades, aircraft Structural Health Monitoring (SHM) technology has gradually turned from fundamental research and laboratorial validations to engineering-oriented developments. However, the process of the transition from research to application has been rather slow and the SHM application to real in-service aircraft structures is barely reported. One of the main application obstacle of the SHM application to real in-service aircraft structures is the problem of reliable damage monitoring under aircraft in-service environmental and operational conditions (referred to as time-varying conditions). Several methods have been proposed to deal with this problem but limitations remain. In this paper, an enhanced Guided Wave-Gaussian Mixture Model (GW-GMM) based damage monitoring method is studied. It can be used on-line without any structural mechanical model or priori knowledge of damage under time-varying conditions. With this method, a baseline GW-GMM is constructed first based on the GW features obtained under time-varying conditions when the structure is in healthy state. When a new GW feature is obtained during an on-line damage monitoring process, the GW-GMM is updated by an enhanced update mechanism including dynamic learning and Gaussian components split-merge. The mixture probability structure of the GW-GMM and the number of Gaussian components can be optimized adaptively. Finally, a Probability Damage Index is used to measure the variation degree between the baseline GW-GMM and the on-line GW-GMM to reveal the weak cumulative variation trend induced by damage of the GW-GMM so as to increase the reliability of damage evaluation. The method is validated in a full-scale aircraft fatigue test and the results indicate that the reliable crack propagation monitoring of the right landing gear spar under the fatigue load condition is achieved