Development of Structural Similarity Metrics for Transfer Learning in Structural Health Monitoring

• Grant number: 24/00720-8
• Support Opportunities: Research Grants - Visiting Researcher Grant - International
• Start date: July 22, 2024
• End date: October 21, 2024
• Field of knowledge: Engineering - Mechanical Engineering - Mechanics of Solids
• Principal Investigator: Samuel da Silva
• Grantee: Samuel da Silva
• Visiting researcher: Eloi Joao Faria Figueiredo
• Visiting researcher institution: Universidade Lusófona de Humanidades e Tecnologias, Portugal
• Host Institution: Faculdade de Engenharia (FEIS). Universidade Estadual Paulista (UNESP). Campus de Ilha Solteira. Ilha Solteira , SP, Brazil

Abstract

Permanent monitoring of infrastructures, such as bridges and dams, is a need for developed societies, as they undergo a process of degradation that can be accelerated by external phenomena. In this way, structural health monitoring (SHM) has been used for detection, diagnosis, and prognosis of damage. The SHM concept has been applied under a statistical pattern recognition paradigm based on artificial intelligence. Recently, transfer learning between similar structures has gained traction, with the advantage of enabling the transfer of structural knowledge between structures. Think about the possibility and impact of using structural responses from the Golden Gate Bridge (1933) in San Francisco to detect damage on the 25 de Abril Bridge in Lisbon (1966). But how to quantify structural similarity, to the point of making this type of technique viable? After several years of collaboration in knowledge transfer, this project aims precisely at developing similarity techniques between structures. In addition to research and publication activities, teaching, guidance, and dissemination activities are also planned. (AU)