Building a machine learning model to predict adhesion force between reinforced concrete and FRP material

Abstract

Fiber-reinforced polymers (FRP) laminate to the external face of reinforced concrete structures is currently one of the most popular and effective methods for strengthening concrete structures. This study aims to compare different machine learning models to predict the adhesion force between reinforced concrete and FRP materials. Single model and ensemble model are used to solve this problem. The analysis results show that the model incorporates packing the regression classification tree - Bagging model (CART) gives the results with the most superior criteria of all the single models and the ensemble models.

Keywords: Adhesion between FRP and concrete; FRP materials; machine learning model.