Prediction of autogenous shrinkage of concretes by support vector machine

Support vector machine (SVM) is firmly based on learning theory and uses regression technique by introducing accuracy insensitive loss function. In this paper, a SVM model for the autogenous shrinkage of concrete mixtures was proposed. The model chose water-to-cementitious material ratio (w/cm), cement content, silica fume percentage, fly ash percentage, total aggregate content, curing temperature, high-range water-reducing admixture (HRWRA) content, and hydration age as input parameters, and the autogenous shrinkage of concrete as the model output. The data set used for training and testing of the SVM model covers the experimental data presented in the existing literature. The developed SVM model was validated using experimental work. The SVM model was compared with the ANN prediction model, the SVM model shows comparable prediction accuracy and could easily be established. In short, the proposed SVM model exhibited excellent capability in predicting the autogenous shrinkage of concrete mixtures.