A series of true triaxial unloading tests are conducted on sandstone specimens with a single structural plane to investigate their mechanical behaviors and failure characteristics under different in situ stress states...A series of true triaxial unloading tests are conducted on sandstone specimens with a single structural plane to investigate their mechanical behaviors and failure characteristics under different in situ stress states.The experimental results indicate that the dip angle of structural plane(θ)and the intermediate principal stress(σ2)have an important influence on the peak strength,cracking mode,and rockburst severity.The peak strength exhibits a first increase and then decrease as a function ofσ2 for a constantθ.However,whenσ2 is constant,the maximum peak strength is obtained atθof 90°,and the minimum peak strength is obtained atθof 30°or 45°.For the case of an inclined structural plane,the crack type at the tips of structural plane transforms from a mix of wing and anti-wing cracks to wing cracks with an increase inσ2,while the crack type around the tips of structural plane is always anti-wing cracks for the vertical structural plane,accompanied by a series of tensile cracks besides.The specimens with structural plane do not undergo slabbing failure regardless ofθ,and always exhibit composite tensile-shear failure whatever theσ2 value is.With an increase inσ2 andθ,the intensity of the rockburst is consistent with the tendency of the peak strength.By analyzing the relationship between the cohesion(c),internal friction angle(φ),andθin sandstone specimens,we incorporateθinto the true triaxial unloading strength criterion,and propose a modified linear Mogi-Coulomb criterion.Moreover,the crack propagation mechanism at the tips of structural plane,and closure degree of the structural plane under true triaxial unloading conditions are also discussed and summarized.This study provides theoretical guidance for stability assessment of surrounding rocks containing geological structures in deep complex stress environments.展开更多
Federated learning is a machine learning framework designed to protect privacy by keeping training data on clients’devices without sharing private data.It trains a global model through collaboration between clients a...Federated learning is a machine learning framework designed to protect privacy by keeping training data on clients’devices without sharing private data.It trains a global model through collaboration between clients and the server.However,the presence of data heterogeneity can lead to inefficient model training and even reduce the final model’s accuracy and generalization capability.Meanwhile,data scarcity can result in suboptimal cluster distributions for few-shot clients in centralized clustering tasks,and standalone personalization tasks may cause severe overfitting issues.To address these limitations,we introduce a federated learning dual optimization model based on clustering and personalization strategy(FedCPS).FedCPS adopts a decentralized approach,where clients identify their cluster membership locally without relying on a centralized clustering algorithm.Building on this,FedCPS introduces personalized training tasks locally,adding a regularization term to control deviations between local and cluster models.This improves the generalization ability of the final model while mitigating overfitting.The use of weight-sharing techniques also reduces the computational cost of central machines.Experimental results on MNIST,FMNIST,CIFAR10,and CIFAR100 datasets demonstrate that our method achieves better personalization effects compared to other personalized federated learning methods,with an average test accuracy improvement of 0.81%–2.96%.Meanwhile,we adjusted the proportion of few-shot clients to evaluate the impact on accuracy across different methods.The experiments show that FedCPS reduces accuracy by only 0.2%–3.7%,compared to 2.1%–10%for existing methods.Our method demonstrates its advantages across diverse data environments.展开更多
基金supports from the National Natural Science Foundation of China (Grant Nos.52004143 and 52374095)the open fund for the Laboratory of Mining Response and Disaster Prevention and Control in Deep Coal Mines (Grant No.SKLMRDPC21KF06).
文摘A series of true triaxial unloading tests are conducted on sandstone specimens with a single structural plane to investigate their mechanical behaviors and failure characteristics under different in situ stress states.The experimental results indicate that the dip angle of structural plane(θ)and the intermediate principal stress(σ2)have an important influence on the peak strength,cracking mode,and rockburst severity.The peak strength exhibits a first increase and then decrease as a function ofσ2 for a constantθ.However,whenσ2 is constant,the maximum peak strength is obtained atθof 90°,and the minimum peak strength is obtained atθof 30°or 45°.For the case of an inclined structural plane,the crack type at the tips of structural plane transforms from a mix of wing and anti-wing cracks to wing cracks with an increase inσ2,while the crack type around the tips of structural plane is always anti-wing cracks for the vertical structural plane,accompanied by a series of tensile cracks besides.The specimens with structural plane do not undergo slabbing failure regardless ofθ,and always exhibit composite tensile-shear failure whatever theσ2 value is.With an increase inσ2 andθ,the intensity of the rockburst is consistent with the tendency of the peak strength.By analyzing the relationship between the cohesion(c),internal friction angle(φ),andθin sandstone specimens,we incorporateθinto the true triaxial unloading strength criterion,and propose a modified linear Mogi-Coulomb criterion.Moreover,the crack propagation mechanism at the tips of structural plane,and closure degree of the structural plane under true triaxial unloading conditions are also discussed and summarized.This study provides theoretical guidance for stability assessment of surrounding rocks containing geological structures in deep complex stress environments.
基金supported by the Foundation of President of Hebei University(XZJJ202303).
文摘Federated learning is a machine learning framework designed to protect privacy by keeping training data on clients’devices without sharing private data.It trains a global model through collaboration between clients and the server.However,the presence of data heterogeneity can lead to inefficient model training and even reduce the final model’s accuracy and generalization capability.Meanwhile,data scarcity can result in suboptimal cluster distributions for few-shot clients in centralized clustering tasks,and standalone personalization tasks may cause severe overfitting issues.To address these limitations,we introduce a federated learning dual optimization model based on clustering and personalization strategy(FedCPS).FedCPS adopts a decentralized approach,where clients identify their cluster membership locally without relying on a centralized clustering algorithm.Building on this,FedCPS introduces personalized training tasks locally,adding a regularization term to control deviations between local and cluster models.This improves the generalization ability of the final model while mitigating overfitting.The use of weight-sharing techniques also reduces the computational cost of central machines.Experimental results on MNIST,FMNIST,CIFAR10,and CIFAR100 datasets demonstrate that our method achieves better personalization effects compared to other personalized federated learning methods,with an average test accuracy improvement of 0.81%–2.96%.Meanwhile,we adjusted the proportion of few-shot clients to evaluate the impact on accuracy across different methods.The experiments show that FedCPS reduces accuracy by only 0.2%–3.7%,compared to 2.1%–10%for existing methods.Our method demonstrates its advantages across diverse data environments.
