Detection of Structural Failures in Aerospace Vehicles via Artificial Intelligence-Based Algorithms

Abstract

Accurate detection of structural damages in aerospace components—such as cracks, corrosion, and manufacturing defects—is of critical importance in the aviation industry. This study aims to enhance the precision and efficiency of damage identification in aerospace structures through the application of artificial intelligence algorithms.

    In this research, data obtained from non-destructive testing (NDT) methods—including ultrasonic testing, radiography, and thermography—were collected and processed using deep learning algorithms. The proposed approach, based on convolutional neural networks (CNNs) and image processing techniques, enabled automated and more accurate damage detection.

   The findings demonstrated that the use of artificial intelligence not only improves diagnostic accuracy compared to traditional methods, but also reduces human error and accelerates data analysis. This approach can serve as an effective tool in periodic inspections of aerospace structures, thereby enhancing their safety and reliability. The results of this study highlight the significant potential of AI in transforming non-destructive inspection techniques.