黑龙江科技大学学报2025，Vol.35Issue(2)：287-293,313,8.DOI:10.3969/j.issn.2095-7262.2025.02.018
基于BO-XGBoost的冲击地压危险等级预测
Prediction of rock burst risk level based on BO-XGBoost
摘要
Abstract
This paper attempts to address the effect of the hyperparameter value on the performance of machine learning model in the prediction of rock burst danger level.The study consists of developing a prediction model of limit gradient lifting(XGBoost)based on Bayesian optimization(BO)algorithm,constructing a data set on the basis of the uniaxial compression experiment on coal and rock mass,optimi-zing the hyperparameters of XGBoost model,Random forest(RF)model and support vector machine(SVM)model by taking acoustic emission signal and compressive strength as the input and danger level as the output,and comparing and analyzing the prediction results.The results show that compared with the traditional XGBoost,RF and SVM,the prediction performance of the three models optimized by BO algorithm is significantly improved.And compared with BO-RF and BO-SVM,the BO-XGBoost has bet-ter generalization ability and higher prediction accuracy of 0.95,0.92 and 0.76,respectively.关键词
冲击地压/声发射/危险等级预测/XGBoost模型/贝叶斯优化Key words
rock burst/acoustic emission/risk level prediction/XGBoost model/Bayes optimiza-tion分类
矿业与冶金引用本文复制引用
李忠勤,宋英才,武俊峰,祝永涛..基于BO-XGBoost的冲击地压危险等级预测[J].黑龙江科技大学学报,2025,35(2):287-293,313,8.基金项目
国家自然科学基金项目(62441306) （62441306）
黑龙江省揭榜挂帅科技攻关项目(2021ZXJ02A02) （2021ZXJ02A02）
