Prediction model of GS-KCV-XGBoost open pit metal mine blasting fragmentation based on feature selection
It's important for mass management to predict accurately and efficiently the blasting blocks.For accurate prediction of blast fragmentation,the crucial factors affecting blasting fragmentation are screened out by random forest algorithm and Pearson correlation analysis,and then input into the XGBoost algorithm optimized by GS and KCV.Therefore,we established an GS-KCV-XGBoost blasting fragmentation prediction model.The research results show that the model has better prediction effect and higher reliability than the common model such as random forest regression model,GS-XGB model,and GS-SVM model.When the model is applied to the actual project,the predicted values are close to the real values,with anR2 of 0.95 and an MAE of 7.961 and an RMSE of 13.59.Therefore,this model can achieve pre-explosion prediction of blasting fragmentation,and has high engineering application value.
NETL
NSTL
万方数据
