Prediction of Unconfined Compressive Strength of Cement Grouting Material Made from Tailings Resource Utilization
Tailings of resource utilization for cement grouting material production contributes to the sustainability of tail-ings management.To accurately predict the unconfined compressive strength of cement grouting material made from tailings,predictive models based on the K-nearest neighbors regression,support vector regression,random forest,Gradient Boosting(GB),and Light Gradient Boosting Machine(LightGBM)algorithms were established.Firstly,a database of unconfined com-pressive strength data for cement grouting material made from tailings was compiled,comprising 738 sets of experimental data.The dataset included parameters such as tailings' chemical properties,cement strength,mass ratio of cement to tailings,mass concentration,curing time,and unconfined compressive strength.Subsequently,predictive models using K-nearest neighbors re-gression,support vector regression,random forest,GB,and LightGBM were established based on this database.Three statistical indicators were selected to evaluate model performance,and model accuracy and errors were compared.The results showed that the ratio of cement to tailings mass exhibits the strongest correlation with unconfined compressive strength,while among the chemical properties of tailings,the silicon dioxide content had the most significant impact on unconfined compressive strength.On both the training and testing datasets,the support vector regression model demonstrated the best predictive performance(with R2 values of approximately 0.99 and 0.98,respectively),with around 99%of the data falling within a 1 MPa range of error.The GB model achieved R2 values of 0.98 and 0.96 on the training and testing datasets,respectively,while the K-nearest neighbors regression model had R2 values of 0.98 and 0.83.Overall,the support vector regression model outperformed the GB model,random forest model,LightGBM model,and K-nearest neighbors regression model,indicating its ability to accurately pre-dict and assess the compressive strength of cement grouting material made from tailings.This research provides a foundation for the resourceful utilization of tailings in the fields of concrete and cement,among others.
tailingsresource utilizationchemical propertiescement grouting materialprediction model
NETL
NSTL
万方数据
