RESUMEN

Guided waves-based SHM systems are of interest in the aeronautic sector due to their lightweight, long interrogation distances, and low power consumption. In this study, a bottom-up framework for the estimation of the initial investment cost (COTC) and the added weight (WAW) associated with the integration of a SHM system to an aircraft is presented. The framework provides a detailed breakdown of the activities and their costs for the sensorization of a structure using a fully wired approach or the adoption of the printed diagnostic film. Additionally, the framework considers the difference between configuring the system for Manual or Remote data acquisition. Based on the case study presented on the sensorization of a regional aircraft composite fuselage, there is a trade-off between COTC and WAW for the SHM options considered. The Wired-Manual case leads to the lowest COTC with the highest WAW, while the combination of diagnostic film with a Remote system leads to the highest COTC and the lowest WAW. These estimations capture the characteristics of each system and can be integrated into cost-benefit analyses for the final selection of a particular configuration.

Asunto(s)

RESUMEN

This paper reports on a novel metamodel for impact detection, localization and characterization of complex composite structures based on Convolutional Neural Networks (CNN) and passive sensing. Methods to generate appropriate input datasets and network architectures for impact localization and characterization were proposed, investigated and optimized. The ultrasonic waves generated by external impact events and recorded by piezoelectric sensors are transferred to 2D images which are used for impact detection and characterization. The accuracy of the detection was tested on a composite fuselage panel which was shown to be over 94%. In addition, the scalability of this metamodelling technique has been investigated by training the CNN metamodels with the data from part of the stiffened panel and testing the performance on other sections with similar geometry. Impacts were detected with an accuracy of over 95%. Impact energy levels were also successfully categorized while trained at coupon level and applied to sub-components with greater complexity. These results validated the applicability of the proposed CNN-based metamodel to real-life application such as composite aircraft parts.