sing the natural limestone samples taken from the field with dimension of 500 mm×500 mm×1 000 mm, the D-D (dilatancy-diffusion) seismogeny pattern was modeled under the condition of water injection, which ob...sing the natural limestone samples taken from the field with dimension of 500 mm×500 mm×1 000 mm, the D-D (dilatancy-diffusion) seismogeny pattern was modeled under the condition of water injection, which observes the time-space evolutionary features about the relative physics fields of the loaded samples from deformation, formation of microcracks to the occurrence of main rupture. The results of observed apparent resistivity show: ① The process of the deformation from microcrack to main rupture on the loaded rock sample could be characterized by the precursory spatial-temporal changes in the observation of apparent resistivity; ② The precursory temporal changes of observation in apparent resistivity could be divided into several stages, and its spatial distribution shows the difference in different parts of the rock sample; ③ Before the main rupture of the rock sample the obvious ″tendency anomaly′ and ′short-term anomaly″ were observed, and some of them could be likely considered as the ″impending earthquake ″anomaly precursor of apparent resistivity. The changes and distribution features of apparent resistivity show that they are intrinsically related to the dilatancy phenomenon of the loaded rock sample. Finally, this paper discusses the mechanism of resistivity change of loaded rock sample theoretically.展开更多
Many empirical and analytical methods have been proposed to predict fracturing pressure in cohesive soils.Most of them take into account three to four specific influencing factors and rely on the assumption of a failu...Many empirical and analytical methods have been proposed to predict fracturing pressure in cohesive soils.Most of them take into account three to four specific influencing factors and rely on the assumption of a failure mode.In this study,a novel data-mining approach based on the XGBoost algorithm is investigated for predicting fracture initiation in cohesive soils.This has the advantage of handling multiple influencing factors simultaneously,without pre-determining a failure mode.A dataset of 416 samples consisting of 14 distinct features was herein collected from past studies,and used for developing a regressor and a classifier model for fracturing pressure prediction and failure mode classification respectively.The results show that the intrinsic characteristics of the soil govern the failure mode while the fracturing pressure is more sensitive to the stress state.The XGBoost-based model was also tested against conventional approaches,as well as a similar machine learning algorithm namely random forest model.Additionally,several large-scale triaxial fracturing tests and an in-situ case study were carried out to further verify the generalization ability and applicability of the proposed data mining approach,and the results indicate a superior performance of the XGBoost model.展开更多
文摘sing the natural limestone samples taken from the field with dimension of 500 mm×500 mm×1 000 mm, the D-D (dilatancy-diffusion) seismogeny pattern was modeled under the condition of water injection, which observes the time-space evolutionary features about the relative physics fields of the loaded samples from deformation, formation of microcracks to the occurrence of main rupture. The results of observed apparent resistivity show: ① The process of the deformation from microcrack to main rupture on the loaded rock sample could be characterized by the precursory spatial-temporal changes in the observation of apparent resistivity; ② The precursory temporal changes of observation in apparent resistivity could be divided into several stages, and its spatial distribution shows the difference in different parts of the rock sample; ③ Before the main rupture of the rock sample the obvious ″tendency anomaly′ and ′short-term anomaly″ were observed, and some of them could be likely considered as the ″impending earthquake ″anomaly precursor of apparent resistivity. The changes and distribution features of apparent resistivity show that they are intrinsically related to the dilatancy phenomenon of the loaded rock sample. Finally, this paper discusses the mechanism of resistivity change of loaded rock sample theoretically.
基金supported by the National Natural Science Foundation of China(Grant No.52008021).
文摘Many empirical and analytical methods have been proposed to predict fracturing pressure in cohesive soils.Most of them take into account three to four specific influencing factors and rely on the assumption of a failure mode.In this study,a novel data-mining approach based on the XGBoost algorithm is investigated for predicting fracture initiation in cohesive soils.This has the advantage of handling multiple influencing factors simultaneously,without pre-determining a failure mode.A dataset of 416 samples consisting of 14 distinct features was herein collected from past studies,and used for developing a regressor and a classifier model for fracturing pressure prediction and failure mode classification respectively.The results show that the intrinsic characteristics of the soil govern the failure mode while the fracturing pressure is more sensitive to the stress state.The XGBoost-based model was also tested against conventional approaches,as well as a similar machine learning algorithm namely random forest model.Additionally,several large-scale triaxial fracturing tests and an in-situ case study were carried out to further verify the generalization ability and applicability of the proposed data mining approach,and the results indicate a superior performance of the XGBoost model.
