Machine Learning for Structural Health Monitoring

• Grant number: 25/09586-5
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: August 01, 2025
• End date: July 31, 2026
• Field of knowledge: Engineering - Mechanical Engineering - Mechanics of Solids
• Principal Investigator: Kayc Wayhs Lopes
• Grantee: Luiz Eduardo Abdala José
• Host Institution: Escola de Engenharia de São Carlos (EESC). Universidade de São Paulo (USP). São Carlos , SP, Brazil

Abstract

This project aims to develop and evaluate machine learning methods for compensating the effects of temperature on signals obtained through the Electromechanical Impedance (EMI) method, as well as for detecting and classifying structural damage. Temperature variation is known to affect EMI signals due to the sensitivity of commonly used sensors and actuators, which can compromise the accuracy of structural diagnostics performed by structural health monitoring systems. To compensate for temperature effects, two distinct machine learning models will be explored: autoencoders and Random Forest. After compensating the EMI curves for temperature effects, the signals will be used for damage detection and classification using decision trees and Random Forest algorithms. The results obtained from the thermally compensated signals will be compared with those from the uncompensated curves to evaluate how temperature compensation influences structural diagnostics based on EMI signals, as well as its impact on the performance of damage detection and classification systems and the computational cost involved in these tasks.