An AIoT Architecture for Structural Testing: Application to a Real Aerospace Component (Embraer E2 Model Aircraft Flag Track)

The AIoT paradigm, which combines AI with IoT, offers great advantages in manufacturing processes. However, its use in aeronautical testing is still incipient, since this kind of test must ensure strict safety requirements. This study presents one AIoT architecture aimed at structurally testing aero...

Full description

Saved in:
Bibliographic Details
Main Authors: Pablo Venegas, Unai Virto, Isidro Calvo, Oscar Barambones
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Applied Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-3417/15/9/4625
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The AIoT paradigm, which combines AI with IoT, offers great advantages in manufacturing processes. However, its use in aeronautical testing is still incipient, since this kind of test must ensure strict safety requirements. This study presents one AIoT architecture aimed at structurally testing aeronautical applications that ease the integration of AI techniques to interpret the data obtained by wireless IoT devices. In addition, the authors propose implementation guidelines for developers. The presented approach was experimentally validated in the rigorous and standardized certification test of a real aerospace component, namely a flag track component of the Embraer E2 model aircraft. Recorded magnitudes with IoT devices were compared with the data obtained using conventional technologies in terms of the quality of information and compliance with the requirements of aeronautical regulations. In order to illustrate the integration of different AI techniques in the AIoT architecture, ARIMA and LSTM algorithms were used to analyze the data captured with three sensors. The obtained results proved that the AIoT architecture is valid in structural testing applications, achieving a reduction in cabling and deployment time as well as improving flexibility and scalability. The presented approach paves the way to introduce AI-based algorithms for analyzing, either in run-time and off-line, the structural testing results obtained by means of IoT devices.
ISSN:2076-3417