About this Event
Reinforced concrete structures rely significantly on the interaction between steel reinforcement bars and the surrounding concrete matrix, referred to as the concrete-steel bond. This bond is critical to the structural integrity and design of reinforced concrete. Predicting its behaviour becomes more complex at elevated temperatures, such as those experienced during fires. An accurate prediction of the bond’s performance is essential for fire-resistant design, yet conventional methods often fall short.
This presentation explores a data-driven approach to predicting concrete-steel bond performance under high temperatures using machine learning (ML) models. Various ML techniques, including Artificial Neural Networks (ANN), Random Forest Regression (RFR), Gradient Boosting (GB), and K-Nearest Neighbours (KNN), are applied to replicate the behaviour of the bond under different thermal conditions. The models are trained on a dataset of 316 data points gathered from laboratory studies testing bond strength at varying temperatures and fibre concentrations.
Early findings show that these ML models effectively capture the complexities of concrete-steel bond behaviour at high temperatures, offering significant improvements over traditional methods. This approach provides engineers a valuable tool for understanding and predicting bond performance, contributing to safer and more resilient infrastructure designs.
Speaker
Dr. Rwayda Al-Hamd
Lecturer in Civil Engineering at Abertay University, UK
Dr Rwayda Al-Hamd is an expert in structural fire safety, specialising in the behaviour of concrete-steel bonds at elevated temperatures. With over ten years of experience in academia and industry, she focuses on applying machine learning models to solve complex engineering challenges. Her research integrates cutting-edge AI technology with structural engineering to enhance the resilience of infrastructure in high-temperature. She is a Fellow of the Higher Education Academy (FHEA) and an active member of the Institution of Structural Engineers (IStructE).
Event Venue
Online
GBP 0.00