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.
Precipitation δ 18O at Yushu, eastern Tibetan Plateau, shows strong fluctuation and lack of clear seasonality. The seasonal pattern of precipitation stable isotope at Yushu is apparently different from either that of the southwest monsoon region to the south or that of the inland region to the north. This different seasonal pattern probably reflects the shift of different moisture sources. In this paper, we present the spatial comparison of the seasonal patterns of precipitation δ 18O, and calculate the moisture transport flux by using the NCAR/NCEP reanalysis data. This allows us to discuss the relation between moisture transport flux and precipitation δ 18O. This study shows that both the southwest monsoon from south and inland air mass transport from north affected the seasonal precipitation δ 18O at Yushu, eastern Tibetan Plateau. Southwest monsoon brings the main part of the moisture, but southwest transport flux is weaker than in the southern part of the Tibetan Plateau. However, contribution of the inland moisture from north or local evaporation moisture is enhanced. The combined effect is the strong fluctuation of summer precipitation δ 18O at Yushu and comparatively poor seasonality. 相似文献
An improvement on the new method of Intervalhalving-Scanning (INS) proposed by the senior author for derivation of thermodynamic properties of minerals from reversed experiment (REP) data has been made in the present work. The treatment of the REP data of 6 reactions in the system MgO-SiO_2-H_2O and the derivation of △_(?)H(?) (298. 15K) for minerals Talc,Forsterite and Anthophyllite are chosen as an example for demonstration of the application of the method.INS is quite different from all the methods for the derivation in the literature and throughout based on thermodynamic principles and equations so that its thermodynamical validity is thoroughgoing. 相似文献
Algae which bloom in open water and accumulate in the littoral zones may affect the biogeochemical cycle of phosphorus in eutrophic lakes. To determine such effects, a part of the lakeshore with little allochthonous nutrient input in Taihu Lake, China was selected for this field study. Distinct differences in sedimentary P forms were found among the different littoral subzones. The surface sedimentary total phosphorus (TP) content was 655 mg/kg in the eulittoral subzone and 631 to 641 mg/kg in the infralittoral subzone. Both were much higher than that in the profundal zone (410 mg/kg). Calcium‐bound P (Ca‐P) was significantly correlated to exchangeable P (Ex‐P), and they both had the highest contents in the infralittoral subzone and the lowest in the profundal zone. The aluminum‐ and iron‐bound P (Al‐P, Fe‐P) contents decreased from land to water along the ecotone section. Lower Fe/P ratios and higher Al‐P/Fe‐P ratios appeared in the infralittoral subzone, as compared with the profundal zone. This suggested that the accumulated algae could lead to a great deposit of P in the littoral zones. However, the active sedimentary P form transformation in the littoral zones would also result in a partial release of the accumulated P to the overlying water. 相似文献