Prediction of Fracture Parameters of Concrete by Relevance Vector Machine

Abstract

This article employs Relevance Vector Machine (RVM) for prediction of two fracture parameters {Critical Stress Intensity Factorand Critical Crack Tip Opening Displacement (CTODc)} of concrete. Input variables of the RVM model are water-cement ratio (), maximum size of aggregate () and characteristic strength (). The developed RVM gives equations for prediction of and CTODc. A comparative study has been presented between the developed RVM and the ANN models. A sensitivity analysis has been done to determine the effect of each input parameter on the predicted and CTODc. This article shows that the developed RVM can be used as an efficient tool for determination of and CTODc.