Using Regression Model to Estimate the Splitting Tensile Strength for the Concrete with Basalt Fiber Reinforced Concrete

Abstract

In this chapter, the optimum model was determined by using Gaussian process, support vector machine, and linear regression models to predict splitting tensile strength of concrete using basalt fiber reinforced polymer (BFRP). For this study, a total of 78 data sets were acquired from various research papers. The whole data set is divided into two subsets: 54 training data sets and 24 testing data sets. The input variables are cement, fine aggregate/crushed sand, coarse aggregate, water, superplasticizer, fly ash, BFRP, diameter of BFRP, length of BFRP, and curing time, whereas the output variable is the compressive strength of the concrete with BFRP. The performance of the generated models is estimated using three performance evaluation indices: root mean square error (RMSE), mean absolute error (MAE), and correlation coefficient (CC). The random forest was shown to be the best model for predicting the compressive strength of concrete with BFRP with CC (0.9942, 0.9170), RMSE (0.0934, 0.3505), and MAE (0.0337, 0.2724) for the training and testing data sets.