Bedding structural planes significantly influence the mechanical properties and stability of engineering rock masses.This study conducts uniaxial compression tests on layered sandstone with various bedding angles(0...Bedding structural planes significantly influence the mechanical properties and stability of engineering rock masses.This study conducts uniaxial compression tests on layered sandstone with various bedding angles(0°,15°,30°,45°,60°,75°and 90°)to explore the impact of bedding angle on the deformational mechanical response,failure mode,and damage evolution processes of rocks.It develops a damage model based on the Logistic equation derived from the modulus’s degradation considering the combined effect of the sandstone bedding dip angle and load.This model is employed to study the damage accumulation state and its evolution within the layered rock mass.This research also introduces a piecewise constitutive model that considers the initial compaction characteristics to simulate the whole deformation process of layered sandstone under uniaxial compression.The results revealed that as the bedding angle increases from 0°to 90°,the uniaxial compressive strength and elastic modulus of layered sandstone significantly decrease,slightly increase,and then decline again.The corresponding failure modes transition from splitting tensile failure to slipping shear failure and back to splitting tensile failure.As indicated by the modulus’s degradation,the damage characteristics can be categorized into four stages:initial no damage,damage initiation,damage acceleration,and damage deceleration termination.The theoretical damage model based on the Logistic equation effectively simulates and predicts the entire damage evolution process.Moreover,the theoretical constitutive model curves closely align with the actual stress−strain curves of layered sandstone under uniaxial compression.The introduced constitutive model is concise,with fewer parameters,a straightforward parameter determination process,and a clear physical interpretation.This study offers valuable insights into the theory of layered rock mechanics and holds implications for ensuring the safety of rock engineering.展开更多
The electric power enterprise is an important basic energy industry for national development,and it is also the first basic industry of the national economy.With the continuous expansion of State Grid,the progressivel...The electric power enterprise is an important basic energy industry for national development,and it is also the first basic industry of the national economy.With the continuous expansion of State Grid,the progressively complex operating conditions,and the increasing scope and frequency of data collection,how to make reasonable use of electrical big data,improve utilization,and provide a theoretical basis for the reliability of State Grid operation,has become a new research hot spot.Since electrical data has the characteristics of large volume,multiple types,low-value density,and fast processing speed,it is a challenge to mine and analyze it deeply,extract valuable information efficiently,and serve for the actual problem.According to the features of these data,this paper uses artificial intelligence methods such as time series and support vector regression to establish a data mining network model for standard cost prediction through transfer learning.The experimental results show that the model in this paper obtains better prediction results on a small sample data set,which verifies the feasibility of the deep transfer model.Compared with activity-based costing and the traditional prediction method,the average absolute error of the proposed method is reduced by 10%,which is effective and superior.展开更多
在网络表示学习的研究中,数据的不完整性问题是一个重要问题,该问题使现有的表示学习算法难以达到预期效果。近年来,不少学者针对此类问题提出了解决方法,这些方法大多仅考虑标签信息本身的缺失问题,对数据不平衡性涉及较少,尤其是某一...在网络表示学习的研究中,数据的不完整性问题是一个重要问题,该问题使现有的表示学习算法难以达到预期效果。近年来,不少学者针对此类问题提出了解决方法,这些方法大多仅考虑标签信息本身的缺失问题,对数据不平衡性涉及较少,尤其是某一类别标签完全缺失的完全不平衡问题。解决这类问题的学习算法并不完善,主要存在的问题是在聚合邻域特征时侧重于考虑网络结构信息,未利用属性特征与语义特征间的关系来增强表示结果。为了解决以上问题,提出了融合属性特征与结构特征的SECT(Semantic Information Enhanced Network Embedding with Completely Imbalanced Labels)方法。首先,在考虑属性空间和语义空间关系的基础上,引入注意力机制进行监督学习,得到语义信息向量;然后,应用变分自编码器无监督提取结构特征以增强算法的鲁棒性;最后,在嵌入空间中融合语义与结构两种信息。将使用SECT算法得到的网络向量表示在Cora,Citeseer等数据集上进行测试,应用于节点分类任务时与RECT和GCN等算法相比,取得了0.86%~1.97%的效果提升。网络向量表示的可视化结果显示,与其他算法相比,SECT算法的类间距离变大,类簇内部更加紧凑,能较清晰地区分类别边界。实验结果表明了SECT算法的有效性,SECT得益于更好地在低维嵌入空间中融合语义信息,有效提升了存在完全不平衡标签情况下的节点分类任务性能。展开更多
基金Projects(52074299,41941018)supported by the National Natural Science Foundation of ChinaProject(2023JCCXSB02)supported by the Fundamental Research Funds for the Central Universities,China。
文摘Bedding structural planes significantly influence the mechanical properties and stability of engineering rock masses.This study conducts uniaxial compression tests on layered sandstone with various bedding angles(0°,15°,30°,45°,60°,75°and 90°)to explore the impact of bedding angle on the deformational mechanical response,failure mode,and damage evolution processes of rocks.It develops a damage model based on the Logistic equation derived from the modulus’s degradation considering the combined effect of the sandstone bedding dip angle and load.This model is employed to study the damage accumulation state and its evolution within the layered rock mass.This research also introduces a piecewise constitutive model that considers the initial compaction characteristics to simulate the whole deformation process of layered sandstone under uniaxial compression.The results revealed that as the bedding angle increases from 0°to 90°,the uniaxial compressive strength and elastic modulus of layered sandstone significantly decrease,slightly increase,and then decline again.The corresponding failure modes transition from splitting tensile failure to slipping shear failure and back to splitting tensile failure.As indicated by the modulus’s degradation,the damage characteristics can be categorized into four stages:initial no damage,damage initiation,damage acceleration,and damage deceleration termination.The theoretical damage model based on the Logistic equation effectively simulates and predicts the entire damage evolution process.Moreover,the theoretical constitutive model curves closely align with the actual stress−strain curves of layered sandstone under uniaxial compression.The introduced constitutive model is concise,with fewer parameters,a straightforward parameter determination process,and a clear physical interpretation.This study offers valuable insights into the theory of layered rock mechanics and holds implications for ensuring the safety of rock engineering.
基金Supported by the program of science and technology of State Grid Zhejiang Electric Power Co.,Ltd.,named Research and application project of standard cost activity based on machine learning(5211JH1900LZ).
文摘The electric power enterprise is an important basic energy industry for national development,and it is also the first basic industry of the national economy.With the continuous expansion of State Grid,the progressively complex operating conditions,and the increasing scope and frequency of data collection,how to make reasonable use of electrical big data,improve utilization,and provide a theoretical basis for the reliability of State Grid operation,has become a new research hot spot.Since electrical data has the characteristics of large volume,multiple types,low-value density,and fast processing speed,it is a challenge to mine and analyze it deeply,extract valuable information efficiently,and serve for the actual problem.According to the features of these data,this paper uses artificial intelligence methods such as time series and support vector regression to establish a data mining network model for standard cost prediction through transfer learning.The experimental results show that the model in this paper obtains better prediction results on a small sample data set,which verifies the feasibility of the deep transfer model.Compared with activity-based costing and the traditional prediction method,the average absolute error of the proposed method is reduced by 10%,which is effective and superior.
文摘在网络表示学习的研究中,数据的不完整性问题是一个重要问题,该问题使现有的表示学习算法难以达到预期效果。近年来,不少学者针对此类问题提出了解决方法,这些方法大多仅考虑标签信息本身的缺失问题,对数据不平衡性涉及较少,尤其是某一类别标签完全缺失的完全不平衡问题。解决这类问题的学习算法并不完善,主要存在的问题是在聚合邻域特征时侧重于考虑网络结构信息,未利用属性特征与语义特征间的关系来增强表示结果。为了解决以上问题,提出了融合属性特征与结构特征的SECT(Semantic Information Enhanced Network Embedding with Completely Imbalanced Labels)方法。首先,在考虑属性空间和语义空间关系的基础上,引入注意力机制进行监督学习,得到语义信息向量;然后,应用变分自编码器无监督提取结构特征以增强算法的鲁棒性;最后,在嵌入空间中融合语义与结构两种信息。将使用SECT算法得到的网络向量表示在Cora,Citeseer等数据集上进行测试,应用于节点分类任务时与RECT和GCN等算法相比,取得了0.86%~1.97%的效果提升。网络向量表示的可视化结果显示,与其他算法相比,SECT算法的类间距离变大,类簇内部更加紧凑,能较清晰地区分类别边界。实验结果表明了SECT算法的有效性,SECT得益于更好地在低维嵌入空间中融合语义信息,有效提升了存在完全不平衡标签情况下的节点分类任务性能。
