Using artificial neural network to predict the fracture properties of the interfacial transition zone of concrete at the meso-scale

Concrete is a multi-phase heterogeneous material in which the interfacial transition zone (ITZ) between aggregates and mortar significantly affects cracking behaviour of concrete, especially under tensile load. In this paper, artificial neural network (ANN) method applied predicting fracture properties ITZ concrete. To form data pool for training ANN, large number two-dimensional (2D) meso-scale simulation concrete direct load conducted. Cohesive crack elements are used simulating arbitrary mortar. After verification trained ANN model, strength energy predicted by using RILEM test results, i.e., stress–displacement curve, as input ANN. It has been found that, performs well computed curve together with optimized parameters good agreement that from results. The randomness little effect on while it becomes slightly bigger energy. ratio to calculated 0.38–0.47 0.18–0.58. These inversely can be useful complementation existing dataset 2D structures.