Prediction of PPV of blasting vibration in open-pit mine based on PCA-WOA-XGBoost model
In order to improve the accuracy of blasting vibration peak speed prediction,a blasting vibration peak speed prediction model based on Whale Optimization Algorithm(WOA)optimized Extreme Gradient Boosting Algorithm(PCA-WOA-XGBoost)under the condition of Principal Component Analysis(PCA)feature dimensionality reduction is proposed.Based on the blasting vibration monitoring data of Changjiu open-pit building materials mine,four principal components were obtained by dimensionality reduction of 11 peak vibration speed influencing factors using principal component analysis,and the scores of the principal components were calculated as the input features of the prediction model,and then the hyper-parameters of the extreme Gradient Boosting Algorithm were optimized using the Whale Algorithm,and the optimal hyper-parameters were inputted to the prediction model for training,testing,and evaluating.The results show that:the dimensionality reduction of the initial features using principal component analysis can effectively reduce the redundancy of information and improve the prediction accuracy;the use of the whale optimization algorithm to optimize the initial hyperparameters of the XGBoost algorithm improves the problem of overfitting caused by manually selecting the hyperparameters;and the average absolute relative error of the prediction results of the PCA-WOA-XGBoost model is 14.59％,which is the lowest among the seven prediction models,and has a higher prediction accuracy.which provides a reference for the prediction of the peak vibration velocity of blasting vibration under the influence of multiple factors.
NETL
NSTL
万方数据
