Prediction of fracture toughness in fibre-reinforced concrete, mortar, and rocks using various machine learning techniques

Abstract

Machine Learning (ML) method is widely used in engineering applications such as fracture mechanics. In this study, twenty different ML algorithms were employed and compared for the prediction of the fracture toughness and fracture load in modes I, II, and mixed-mode (I-II) of various materials, including fibre-reinforced concrete, cement mortar, sandstone, white travertine, marble, and granite. A set of 401 specimens of “Brazilian discs with central cracks” were used as a training and testing dataset. The main features of the experimental technique in each specimen are the fracture mode, the tensile strength of the specimen, the inclination of the crack with loading direction, the thickness of specimens and the half-length of the crack. The improved ML algorithms were implemented using Python programming language. The results of the coefficient of restitution (R2) and statistical metrics confirm that the ML algorithms are able to predict the fracture toughness and fracture load in modes I, II, and mixed-mode (I-II) with high accuracy. To validate the reliability of the proposed ML-based prediction models, three experimental tests were used. Moreover, the Graphical User Interface (GUI) of the ML-based models was created as a practical tool for estimating the fracture load and fracture toughness for engineering problems.