Damage Detection in Aeronautic Structures using Piezoelectric Sensors and Machine Learning Models

The widespread adoption of composite materials in the aeronautic industry has generated a critical need for effective structural health monitoring (SHM) to ensure reliability and safety. The integration of smart composite structures with embedded sensors has emerged as a promising approach for detecting and assessing structural damage through the analysis of vibration responses across a range of frequencies. This study investigates the application of machine learning models for damage detection in aeronautic composite structures, focusing on their ability to analyze complex vibration signals and distinguish between healthy and damaged states. An experimental framework is presented to evaluate the performance of various machine learning models in identifying faults under different vibration conditions and fault types. Key research questions addressed include: (1) Can machine learning models reliably detect damage in smart composite structures? (2) Which models exhibit superior performance across varying operational conditions, or do they demonstrate significant variability in robustness across frequency domains? The findings offer critical insights into the comparative performance of machine learning algorithms in the context of structural damage detection. This research contributes to the advancement of data-driven SHM techniques, enhancing the reliability and efficiency of monitoring systems in aeronautic applications.