Fault Detection and Prognosis in Carbon Fiber Reinforced Polymers Using Acoustic Emission, Machine Learning, and Computer Vision

• Grant number: 25/06677-0
• Support Opportunities: Scholarships in Brazil - Master
• Start date: January 01, 2026
• End date: February 28, 2027
• Field of knowledge: Engineering - Electrical Engineering
• Principal Investigator: Pedro de Oliveira Conceição Junior
• Grantee: Catherine Bezerra Markert
• Host Institution: Escola de Engenharia de São Carlos (EESC). Universidade de São Paulo (USP). São Carlos , SP, Brazil
• Associated research grant: 23/02413-2 - Smart systems for fault diagnosis and monitoring of industrial assets based on the industry 4.0 concept, AP.R

Abstract

This project is related to the Regular Research Grant funded by FAPESP (Proc. 2023/02413-2), entitled Intelligent systems for monitoring and diagnosing failures in industrial assets based on the Industry 4.0 concept. Carbon Fiber Reinforced Polymers (CFRPs) have been increasingly used in various fields, including automotive, energy, sports, and biomedical, due to their high mechanical strength and low weight. However, these materials are susceptible to delamination, which can compromise their structural integrity and lead to failure. In this context, the project proposes the analysis of Acoustic Emission (AE) signals, a technique widely used in damage monitoring, with the goal of identifying and characterizing failures in composites. The project aims to deepen the understanding of degradation mechanisms in CFRPs from AE signals, also correlating them with tool wear during machining. Machine learning models will be developed to estimate faults, such as delamination and tool wear, while computer vision will assist in identifying patterns related to damage evolution and the tool's remaining useful life (RUL). The study will be validated through experimental analysis involving the collection of AE signals and corresponding microscopic photographs of different severity of structural faults in composite materials. The obtained results are expected to contribute to more precise methods of structural monitoring and to the scientific advancement in understanding composite degradation in machining processes.