In the mining industry,precise forecasting of rock fragmentation is critical for optimising blasting processes.In this study,we address the challenge of enhancing rock fragmentation assessment by developing a novel hy...In the mining industry,precise forecasting of rock fragmentation is critical for optimising blasting processes.In this study,we address the challenge of enhancing rock fragmentation assessment by developing a novel hybrid predictive model named GWO-RF.This model combines the grey wolf optimization(GWO)algorithm with the random forest(RF)technique to predict the D_(80)value,a critical parameter in evaluating rock fragmentation quality.The study is conducted using a dataset from Sarcheshmeh Copper Mine,employing six different swarm sizes for the GWO-RF hybrid model construction.The GWO-RF model’s hyperparameters are systematically optimized within established bounds,and its performance is rigorously evaluated using multiple evaluation metrics.The results show that the GWO-RF hybrid model has higher predictive skills,exceeding traditional models in terms of accuracy.Furthermore,the interpretability of the GWO-RF model is enhanced through the utilization of SHapley Additive exPlanations(SHAP)values.The insights gained from this research contribute to optimizing blasting operations and rock fragmentation outcomes in the mining industry.展开更多
Forest fire accidents caused by distribution line faults occur frequently,resulting in heavy impacts on people’s safety and social and economic development.Currently,there are few risk assessments for forest fires in...Forest fire accidents caused by distribution line faults occur frequently,resulting in heavy impacts on people’s safety and social and economic development.Currently,there are few risk assessments for forest fires induced by over-head distribution lines,and existing assessment methods may have difficulties in data acquisition.On this basis,a novel as-sessment framework based on an analytic hierarchy process,a Bayesian network and a Fussel-Vesely importance metric is proposed in this paper.The framework combines field research and historical operation and maintenance data to assess the regional-scale risk of forest fires induced by overhead distribution lines to derive the probability of forest fires and to identify high-risk lines and key hazard events in the assessment region.Finally,taking the southern Anhui region as an ex-ample,the annual fire probability of forest fires induced by overhead distribution lines in the southern Anhui region is 5.88%,and rectification measures are proposed.This study provides management with a complete assessment framework that optimizes the difficulty of data collection and allows for additional targeted corrective measures to be proposed for the entire region and route on the basis of the assessment results.展开更多
Using flexible damping technology to improve tunnel lining structure is an emerging method to resist earthquake disasters,and several methods have been explored to predict mechanical response of tunnel lining with dam...Using flexible damping technology to improve tunnel lining structure is an emerging method to resist earthquake disasters,and several methods have been explored to predict mechanical response of tunnel lining with damping layer.However,the traditional numerical methods suffer from the complex modelling and time-consuming problems.Therefore,a prediction model named the random forest regressor(RFR)is proposed based on 240 numerical simulation results of the mechanical response of tunnel lining.In addition,circle mapping(CM)is used to improve Archimedes optimization algorithm(AOA),reptile search algorithm(RSA),and Chernobyl disaster optimizer(CDO)to further improve the predictive performance of the RFR model.The performance evaluation results show that the CMRSA-RFR is the best prediction model.The damping layer thickness is the most important feature for predicting the maximum principal stress of tunnel lining containing damping layer.This study verifies the feasibility of combining numerical simulation with machine learning technology,and provides a new solution for predicting the mechanical response of aseismic tunnel with damping layer.展开更多
The residual elastic energy index is a scientific evaluation index for rockburst proneness.In laboratory test,it is sometimes difficult to obtain the post-peak curve or to test the rock sample several times,which make...The residual elastic energy index is a scientific evaluation index for rockburst proneness.In laboratory test,it is sometimes difficult to obtain the post-peak curve or to test the rock sample several times,which makes it impossible to calculate the residual elastic energy index accurately.Based on 241 sets of experimental data and four input indexes of density,elastic modulus,peak intensity and peak input strain energy,this study proposed a machine learning model combining k-means clustering algorithm and random forest regression model:cluster forest(CF)model.The research employed a stratified sampling method on the dataset to ensure the representativeness and balance of the samples.Subsequently,grid search and five-fold cross-validation were utilized to optimize the model’s hyperparameters,aiming to enhance its generalization capability and prediction accuracy.Finally,the performance of the optimal model was evaluated using a test set and compared with five other commonly used models.The results indicate that the CF model outperformed the other models on the testing set,with a mean absolute error of 6.6%,and an accuracy of 93.9%.The results of sensitivity analyses reveal the degree of influence of each variable on rockburst proneness and the applicability of the CF model when the input parameters are missing.The robustness and generalization ability of the model were verified by introducing experimental data from other studies,and the results confirmed the reliability and applicability of the model.Therefore,the model not only effectively simplifies the acquisition of the residual elastic energy index,but also shows excellent performance and wide applicability.展开更多
Underground excavation can lead to stress redistribution and result in an excavation damaged zone(EDZ),which is an important factor affecting the excavation stability and support design.Accurately estimating the thick...Underground excavation can lead to stress redistribution and result in an excavation damaged zone(EDZ),which is an important factor affecting the excavation stability and support design.Accurately estimating the thickness of EDZ is essential to ensure the safety of the underground excavation.In this study,four novel hybrid ensemble learning models were developed by optimizing the extreme gradient boosting(XGBoost)and random forest(RF)algorithms through simulated annealing(SA)and Bayesian optimization(BO)approaches,namely SA-XGBoost,SA-RF,BO XGBoost and BO-RF models.A total of 210 cases were collected from Xiangxi Gold Mine in Hunan Province and Fankou Lead-zinc Mine in Guangdong Province,China,including seven input indicators:embedding depth,drift span,uniaxial compressive strength of rock,rock mass rating,unit weight of rock,lateral pressure coefficient of roadway and unit consumption of blasting explosive.The performance of the proposed models was evaluated by the coefficient of determination,root mean squared error,mean absolute error and variance accounted for.The results indicated that the SA-XGBoost model performed best.The Shapley additive explanations method revealed that the embedding depth was the most important indicator.Moreover,the convergence curves suggested that the SA-XGBoost model can reduce the generalization error and avoid overfitting.展开更多
基金Projects(42177164,52474121)supported by the National Science Foundation of ChinaProject(PBSKL2023A12)supported by the State Key Laboratory of Precision Blasting and Hubei Key Laboratory of Blasting Engineering,China。
文摘In the mining industry,precise forecasting of rock fragmentation is critical for optimising blasting processes.In this study,we address the challenge of enhancing rock fragmentation assessment by developing a novel hybrid predictive model named GWO-RF.This model combines the grey wolf optimization(GWO)algorithm with the random forest(RF)technique to predict the D_(80)value,a critical parameter in evaluating rock fragmentation quality.The study is conducted using a dataset from Sarcheshmeh Copper Mine,employing six different swarm sizes for the GWO-RF hybrid model construction.The GWO-RF model’s hyperparameters are systematically optimized within established bounds,and its performance is rigorously evaluated using multiple evaluation metrics.The results show that the GWO-RF hybrid model has higher predictive skills,exceeding traditional models in terms of accuracy.Furthermore,the interpretability of the GWO-RF model is enhanced through the utilization of SHapley Additive exPlanations(SHAP)values.The insights gained from this research contribute to optimizing blasting operations and rock fragmentation outcomes in the mining industry.
基金This work was supported by the National Key Research and Development Program of China(2022YFC3003101)the Fundamental Research Funds for the Central Universities(WK2320000050)the Science and Technology Program of State Grid Anhui Electric Power Co.,Ltd.(521205220001).
文摘Forest fire accidents caused by distribution line faults occur frequently,resulting in heavy impacts on people’s safety and social and economic development.Currently,there are few risk assessments for forest fires induced by over-head distribution lines,and existing assessment methods may have difficulties in data acquisition.On this basis,a novel as-sessment framework based on an analytic hierarchy process,a Bayesian network and a Fussel-Vesely importance metric is proposed in this paper.The framework combines field research and historical operation and maintenance data to assess the regional-scale risk of forest fires induced by overhead distribution lines to derive the probability of forest fires and to identify high-risk lines and key hazard events in the assessment region.Finally,taking the southern Anhui region as an ex-ample,the annual fire probability of forest fires induced by overhead distribution lines in the southern Anhui region is 5.88%,and rectification measures are proposed.This study provides management with a complete assessment framework that optimizes the difficulty of data collection and allows for additional targeted corrective measures to be proposed for the entire region and route on the basis of the assessment results.
基金Project(2023YFB2390400)supported by the National Key R&D Programs for Young Scientists,ChinaProjects(U21A20159,52079133,52379112,52309123,41902288)supported by the National Natural Science Foundation of China+5 种基金Project(2024AFB041)supported by the Hubei Provincial Natural Science Foundation,ChinaProject(QTKS0034W23291)supported by the Key Laboratory of Water Grid Project and Regulation of Ministry of Water Resources,ChinaProject(2023SGG07)supported by the Visiting Researcher Fund Program of State Key Laboratory of Water Resources Engineering and Management,ChinaProject(2022KY56(ZDZX)-02)supported by the Key Research Program of FSDI,ChinaProject(SKS-2022103)supported by the Key Research Program of the Ministry of Water Resources,ChinaProject(202102AF080001)supported by the Yunnan Major Science and Technology Special Program,China。
文摘Using flexible damping technology to improve tunnel lining structure is an emerging method to resist earthquake disasters,and several methods have been explored to predict mechanical response of tunnel lining with damping layer.However,the traditional numerical methods suffer from the complex modelling and time-consuming problems.Therefore,a prediction model named the random forest regressor(RFR)is proposed based on 240 numerical simulation results of the mechanical response of tunnel lining.In addition,circle mapping(CM)is used to improve Archimedes optimization algorithm(AOA),reptile search algorithm(RSA),and Chernobyl disaster optimizer(CDO)to further improve the predictive performance of the RFR model.The performance evaluation results show that the CMRSA-RFR is the best prediction model.The damping layer thickness is the most important feature for predicting the maximum principal stress of tunnel lining containing damping layer.This study verifies the feasibility of combining numerical simulation with machine learning technology,and provides a new solution for predicting the mechanical response of aseismic tunnel with damping layer.
基金Project(42077244)supported by the National Natural Science Foundation of ChinaProject(SDGZK2431)supported by the State Key Laboratory of Intelligent Construction and Healthy Operation and Maintenance of Deep Underground Engineering,Sichuan University,China。
文摘The residual elastic energy index is a scientific evaluation index for rockburst proneness.In laboratory test,it is sometimes difficult to obtain the post-peak curve or to test the rock sample several times,which makes it impossible to calculate the residual elastic energy index accurately.Based on 241 sets of experimental data and four input indexes of density,elastic modulus,peak intensity and peak input strain energy,this study proposed a machine learning model combining k-means clustering algorithm and random forest regression model:cluster forest(CF)model.The research employed a stratified sampling method on the dataset to ensure the representativeness and balance of the samples.Subsequently,grid search and five-fold cross-validation were utilized to optimize the model’s hyperparameters,aiming to enhance its generalization capability and prediction accuracy.Finally,the performance of the optimal model was evaluated using a test set and compared with five other commonly used models.The results indicate that the CF model outperformed the other models on the testing set,with a mean absolute error of 6.6%,and an accuracy of 93.9%.The results of sensitivity analyses reveal the degree of influence of each variable on rockburst proneness and the applicability of the CF model when the input parameters are missing.The robustness and generalization ability of the model were verified by introducing experimental data from other studies,and the results confirmed the reliability and applicability of the model.Therefore,the model not only effectively simplifies the acquisition of the residual elastic energy index,but also shows excellent performance and wide applicability.
基金Project(52204117)supported by the National Natural Science Foundation of ChinaProject(2022JJ40601)supported by the Natural Science Foundation of Hunan Province,China。
文摘Underground excavation can lead to stress redistribution and result in an excavation damaged zone(EDZ),which is an important factor affecting the excavation stability and support design.Accurately estimating the thickness of EDZ is essential to ensure the safety of the underground excavation.In this study,four novel hybrid ensemble learning models were developed by optimizing the extreme gradient boosting(XGBoost)and random forest(RF)algorithms through simulated annealing(SA)and Bayesian optimization(BO)approaches,namely SA-XGBoost,SA-RF,BO XGBoost and BO-RF models.A total of 210 cases were collected from Xiangxi Gold Mine in Hunan Province and Fankou Lead-zinc Mine in Guangdong Province,China,including seven input indicators:embedding depth,drift span,uniaxial compressive strength of rock,rock mass rating,unit weight of rock,lateral pressure coefficient of roadway and unit consumption of blasting explosive.The performance of the proposed models was evaluated by the coefficient of determination,root mean squared error,mean absolute error and variance accounted for.The results indicated that the SA-XGBoost model performed best.The Shapley additive explanations method revealed that the embedding depth was the most important indicator.Moreover,the convergence curves suggested that the SA-XGBoost model can reduce the generalization error and avoid overfitting.
