Prediction of Concrete Strength Based on Random Forest and Gradient Boosting Machine

Abstract

The design of concrete proportion is directly related to the engineering performance and cost of concrete, and the conventional concrete proportion design method often requires a lot of trial-formulation practice, spending a lot of time and money to get a concrete proportion that meets the requirements. In this paper, it is proposed to use the technology of artificial intelligence to establish the prediction model of the compressive strength of concrete for the corresponding mix ratio to assist in the trial-formulation work in order to reduce the cost in the trial-formulation work.Based on regression model, especially Random Forest regression model and Gradient Boosting Machine, this paper uses 8 properties of cement, water and coarse aggregate to predict the compressive strength corresponding to the concrete mix proportion. The R-squared of the Random Forest regression model reached 0.9167, and the R-squared of the Gradient Boosting Machine reached 0.942.