Upcoming damage size quantification in aeronautic composite structures based on imaging results post-processing

Abstract

In this paper, a damage quantification strategy relying on post-processing of Lamb wave based damage localization results is presented. This method is able to predict the upcoming sizes of a delamination after a training step. Inputs of the proposed method are localization index maps produced by damage localization algorithms and representing the presence likelihood of a damage over the structure under study. The area covered by a high localization index around the estimated damage location are then extracted from these spatial probability maps. A data-driven model representing the mathematical relationship between this quantification feature and the actual size of the damage is finally inferred and used to predict future damage size. The proposed method is successfully validated on experimental data coming from CFRP plate samples equipped with piezoelectric transducers. Delaminations induced by fatigue testing and laser shock are studied. The sensitivity of the method to input frequency and damage localization algorithms parameters is assessed and a method to automatically select its own parameters is proposed. Furthermore, it is demonstrated that a model can be confidently learned on a given CFRP plate sample and transferred to predict damage size on another similar CFRP plate sample.

Image processing through deep learning after DI extraction for the SHM of aeronautic composite structures using Lamb waves

“…After each signal has been collected for all receiver-actuator pairs, damaged states signals are compared to the reference signals in order to compute 40 candidate Damage Indexes (DI) designed for damage detection. In section 5.2 of chapter 4 of his PhD thesis [8], Briand studied families of DI on several test plates each with a different fatigue scenario (delamination, impact). The objective was to compute an average score of 3 sub-scores determining the coherence, the range, and the consistency of the DIs when faced with different scenarios.…”

Cross-structures Deep Transfer Learning through Kantorovich potentials for Lamb Waves based Structural Health Monitoring.

“…Damage classification, is harder to achieve and to test in practice as several damage types are needed, which is complicated to manage in practice for composite materials [9,13]. Finally, the damage quantification step is the most delicate one and is a current area of research [14,15]. Consequently, the focus is here put on the damage localization and severity quantification steps of the LWSHM process.…”

The Need for Multi-Sensor Data Fusion in Structural Health Monitoring of Composite Aircraft Structures

“…Note that the proposed framework is discussed on a high level in a conceptual manner, and merely serves to demonstrate the first indications of the enhanced capabilities of a multi-sensor-based SHM system. To fully unleash its potential, in-depth research studies are crucial that consist of experimental campaigns, detailed diagnostic and prognostic methodology development, and validation exercises [8,36,50,72,[89][90][91][92][93][94][95][96][97][98], which are beyond the scope of this work.…”