Application of the Elastic Waves and Neural Networks as a Tool of Damage Detection and Health Monitoring in Aircraft's Structures

There exist a large group of structures like e.g. aircraft, which operational safety requires periodical inspections or even continuous monitoring of its health. Sometimes the structures are equipped with Structural Health Monitoring systems which usually utilizes the phenomena of elastic waves propagation in solids. The analysis of elastic waves signals consists in general of quantitative and qualitative description of theirs changes (e.g. attenuation, distortion, reflection) caused by a damage appearance and growth. Since the reflections and dispersion effects may produce pretty complex signals, therefore the determination of parameters suitable for damage detection requires the application of advance signal processing techniques. For this purpose an approach of novelty detection and damage evaluation based on soft computing methods (for example - Neural Networks) was proposed in this paper. Two levels of the damage identification problem were realized: novelty detection and damage assessment. The system accuracy and reliability were verified during laboratory tests. It was proved that the system can be used for the analysis of simple as well as complex signals. One of the important factors in the structural health monitoring systems is the amount of data that need to be analysed in real time. This study investigated the use of artificially deteriorated signals of elastic waves in training the novelty detection (ND) system. In this system auto-associative neural networks were trained using the principal components calculated on the basis of experimentally measured signals. It was found that the designed ND system remained sensitive and robust even when it uses raw signals with a relatively low sampling rate, on a fairly narrow time window and even noisy signals.