Rockburst is a common dynamic geological hazard, severely restricting the development and utilization of underground space and resources. As the depth of excavation and mining increases, rockburst tends to occur frequently. Hence, it is necessary to carry out a study on rockburst prediction. Due to the nonlinear relationship between rockburst and its influencing factors, artificial intelligence was introduced. However, the collected data were typically imbalanced. Single algorithms trained by such data have low recognition for minority classes. In order to handle the problem, this paper employed stacking technique of ensemble learning to establish rockburst prediction models. In total, 246 sets of data were collected. In the preprocessing stage, three data mining techniques including principal component analysis, local outlier factor and expectation maximization algorithm were used for dimension reduction, outlier detection and outlier substitution, respectively. Then, the pre-processed data were split into a training set (75%) and a test set (25%) with stratified sampling. Based on the four classical single intelligent algorithms, namely k-nearest neighbors (KNN), support vector machine (SVM), deep neural network (DNN) and recurrent neural network (RNN), four ensemble models (KNN–RNN, SVM–RNN, DNN–RNN and KNN–SVM–DNN–RNN) were built by stacking technique of ensemble learning. The prediction performance of eight models was evaluated, and the differences between single models and ensemble models were analyzed. Additionally, a sensitivity analysis was conducted, revealing the importance of input variables on the models. Finally, the impact of class imbalance on the prediction accuracy and fitting effect of models was quantitatively discussed. The results showed that stacking technique of ensemble learning provides a new and promising way for rockburst prediction, which exhibits unique advantages especially when using imbalanced data.
Natural Resources Research - In the process of open-pit bench blasting for many mines, rock fragments move in the direction of loose space after fragmentation under explosive energy, leading to ore... 相似文献
Regional ecological health,the core of comprehensive ecosystem assessments,is an important foundation for regional exploration,environmental conservation,and sustainable development.The mountainous areas in southwest China are backward in economy,but industrialization and urbanization have been rapid in recent years.This study assessed the ecosystem health of the Sichuan and Yunnan provinces in China using a pressure-state-response(PSR)model.Spatiotemporal patterns of regional ecosystem health were analyzed from 2000 to 2016,including overall characteristics as well as local characteristics.Ecosystem health in most regions was improved over time(Y=0.0058 X–11.0132,R2=0.95,P<0.001),and areas with poorer ecosystem health decreased from half to one-third of the total area.Analysis of the primacy ratio and the variation coefficient confirmed that the gap in health scores between regions has gradually expanded since 2007,but there are more high quality regions overall(Z of Moran’s index<1.96,P>0.05).Overall,the regional ecosystems to the east of the Hu line-an imaginary line dividing east and west China into roughly equivalent parts-were healthier than those to the west.The pressure and state scores of ecosystems were determined by physiographic condition,and the response scores by government policies and social concern.The spatiotemporal patterns of ecosystem health were dominated to a greater extent by natural than anthropogenic factors,which explains why the shift in the patterns aligned with the direction of the Hu line.Dividing regions into key management areas based on natural geographical conditions and socioeconomic development could contribute to the formulation of a reasonable ecological and environmental protection policy,guaranteeing ecosystem services in the long run. 相似文献
