Prediction of High Performance Concrete Compressive Strength based on Stacking
In order to achieve rapid and accurate prediction of the compressive strength of high perform-ance concrete(HPC),the study is based on the dataset of 58 HPC mix proportions,utilizing Stacking ensemble learning model with nine interpretable features as input variables to predict the compressive strength of HPC compared with the other four single models.The results showed that compared with the traditional base models,the Stacking ensemble learning model had the smallest error values and the largest determination coefficient.The mean absolute percentage error,mean absolute error,root mean square error and correlation coefficients for compressive strength prediction of the high performance concrete were 11.40％、3.72、5.04、0.91,respectively,The proposed model had higher accuracy in pre-dicting HPC compressive strength.
high performance concretecompressive strengthbase modelStacking
NETL
NSTL
万方数据
