Determination of Compressive Strength of Concrete by Statistical Learning Algorithms

Abstract

This article adopts three statistical learning algorithms: support vector machine (SVM), lease square support vector machine (LSSVM), and relevance vector machine (RVM), for predicting compressive strength (f_c) of concrete. Fly ash replacement ratio (FA), silica fume replacement ratio (SF), total cementitious material (TCM), fine aggregate (ssa), coarse aggregate (ca), water content (W), high rate water reducing agent (HRWRA), and age of samples (AS) are used as input parameters of SVM, LSSVM and RVM. The output of SVM, LSSVM and RVM is f_c. This article gives equations for prediction of f_c of concrete. A comparative study has been carried out between the developed SVM, LSSVM, RVM and Artificial Neural Network (ANN). This article shows that the developed SVM, LSSVM and RVM models are practical tools for the prediction of f_c of concrete.