Early Detection of Concrete Reinforcement Corrosion using Nonlinear Ultrasonic Techniques and Artificial Neural Networks

• Grant number: 23/17255-3
• Support Opportunities: Scholarships in Brazil - Doctorate
• Start date: August 01, 2024
• End date: May 31, 2028
• Field of knowledge: Engineering - Civil Engineering - Construction Industry
• Principal Investigator: Vladimir Guilherme Haach
• Grantee: Lara Guizi Anoni
• Host Institution: Escola de Engenharia de São Carlos (EESC). Universidade de São Paulo (USP). São Carlos , SP, Brazil

Abstract

Corrosion of reinforcement in concrete structures presents a significant challenge, impacting the long-term performance and safety of these constructions. Early detection of corrosion is crucial for cost-effective maintenance and the prevention of structural collapse. Various Non-Destructive Testing (NDT) techniques, particularly ultrasonic methods, show promise in early corrosion detection, including the assessment of microcracks and debonding. Nevertheless, interpreting the results of these tests remains complex. To address the challenge, this research proposes a novel methodology that employs Artificial Neural Networks (ANNs) along with ultrasonic testing for corrosion detection. This approach explores the application of ultrasonic tomographic images and Nonlinear Ultrasonic (NLU) evaluation for signal analysis. The expected results include enhanced corrosion and structural assessment efficiency, improved maintenance decision-making, and the promotion of structural longevity and safety.