该文将深度学习用于遥感图像融合,在训练深度网络时加入了结构风险最小化的损失函数,提出了一种基于深度支撑值学习网络的融合方法.为了避免图像融合过程中的信息损失,在传统卷积神经网络的基础上,取消了特征映射层的下采样过程,构建了...该文将深度学习用于遥感图像融合,在训练深度网络时加入了结构风险最小化的损失函数,提出了一种基于深度支撑值学习网络的融合方法.为了避免图像融合过程中的信息损失,在传统卷积神经网络的基础上,取消了特征映射层的下采样过程,构建了深度支撑值学习网络(Deep Support Value Learning Networks,DSVL Nets),DSVL Nets网络模型包含5个隐藏层,每一层的基本结构由卷积层和线性层构成,该基本单元提供了一种多尺度、多方向、各向异性、非下采样的冗余变换,该模型在网络训练完毕之后,取出各卷积层和第5个隐藏层的线性层作为网络模型的输出层.输出层的各卷积层图像融合采用绝对值取大法,得到融合后的各卷积层图像;另外,将线性层图像分别在过完备字典上进行稀疏表示,并对稀疏系数采用绝对值取大法进行融合,得到融合后的线性层图像;最后将融合后的各卷积层和线性层图像重构得到结果图像.文中使用QuickBird和Geoeye卫星数据验证本文所提方法的有效性,实验结果表明,与PCA、AWLP、PN-TSSC和SVT算法相比较,该文所提方法的融合结果无论在主观视觉还是客观评价指标上均优于对比算法,较好地保持了图像的光谱信息和空间信息.展开更多
Based on nonlinear Mohr-Coulomb failure criterion, the analytical solutions of stability number and supporting force on twin shallow tunnels were derived using upper bound theorem of limit analysis. The optimized solu...Based on nonlinear Mohr-Coulomb failure criterion, the analytical solutions of stability number and supporting force on twin shallow tunnels were derived using upper bound theorem of limit analysis. The optimized solutions were obtained by the technique of sequential quadratic programming. When nonlinear coefficient equals 1 and internal friction angle equals 0, the nonlinear Mohr-Coulomb failure criterion degenerates into linear failure criterion. The calculated results of stability number in this work were compared with previous results, and the agreement verifies the effectiveness of the present method. Under the condition of nonlinear Mohr-Coulomb failure criterion, the results show that the supporting force on twin shallow tunnels obviously increases when the nonlinear coefficient, burial depth, ground load or pore water pressure coefficients increase. When the clear distance is 0.5to 1.0 times the diameter of tunnel, the supporting force of twin shallow tunnels reaches its maximum value, which means that the tunnels are the easiest to collapse. While the clear distance increases to 3.5 times the diameter of tunnel, the calculation for twin shallow tunnels can be carried out by the method for independent single shallow tunnel. Therefore, 3.5 times the diameter of tunnel serves as a critical value to determine whether twin shallow tunnels influence each other. In designing twin shallow tunnels,appropriate clear distance value must be selected according to its change rules and actual topographic conditions, meanwhile, the influences of nonlinear failure criterion of soil materials and pore water must be completely considered. During the excavation process, supporting system should be intensified at the positions of larger burial depth or ground load to avoid collapses.展开更多
Hydraulic fracturing is designed to form a high-conductivity fracture. The proppant will embed into the formation rock, especially in soft rock, owing to the interaction between proppant and fracture surface after fra...Hydraulic fracturing is designed to form a high-conductivity fracture. The proppant will embed into the formation rock, especially in soft rock, owing to the interaction between proppant and fracture surface after fracture closure. Proppant embedment would reduce the fracture width and then lower the fracture conductivity. According to dimensional analysis, the rock is assumed to be an elastic material. Using the theory of elasticity to describe the stage of elastic deformation and analysis of the corresponding simplified embedding process, the study establish the static computation model of the two-dimensional infinite half plane and three-dimensional infinite half space model of the proppant embedment. According to laboratory results, the calculation model was modified, got an effective correction factor and analyzed the causes of errors, then discussed the factors which have impact on proppant embedment. The result calculated by the model in this paper can be reference of prop- pant optimization in on-site fracturing for a certainty degree.展开更多
文摘该文将深度学习用于遥感图像融合,在训练深度网络时加入了结构风险最小化的损失函数,提出了一种基于深度支撑值学习网络的融合方法.为了避免图像融合过程中的信息损失,在传统卷积神经网络的基础上,取消了特征映射层的下采样过程,构建了深度支撑值学习网络(Deep Support Value Learning Networks,DSVL Nets),DSVL Nets网络模型包含5个隐藏层,每一层的基本结构由卷积层和线性层构成,该基本单元提供了一种多尺度、多方向、各向异性、非下采样的冗余变换,该模型在网络训练完毕之后,取出各卷积层和第5个隐藏层的线性层作为网络模型的输出层.输出层的各卷积层图像融合采用绝对值取大法,得到融合后的各卷积层图像;另外,将线性层图像分别在过完备字典上进行稀疏表示,并对稀疏系数采用绝对值取大法进行融合,得到融合后的线性层图像;最后将融合后的各卷积层和线性层图像重构得到结果图像.文中使用QuickBird和Geoeye卫星数据验证本文所提方法的有效性,实验结果表明,与PCA、AWLP、PN-TSSC和SVT算法相比较,该文所提方法的融合结果无论在主观视觉还是客观评价指标上均优于对比算法,较好地保持了图像的光谱信息和空间信息.
基金Project(2013CB036004)supported by the National Basic Research Program of ChinaProject(51378510)supported by the NationalNatural Science Foundation of ChinaProject(CX2013B077)supported by Hunan Provincial Innovation Foundation for Postgraduate,China
文摘Based on nonlinear Mohr-Coulomb failure criterion, the analytical solutions of stability number and supporting force on twin shallow tunnels were derived using upper bound theorem of limit analysis. The optimized solutions were obtained by the technique of sequential quadratic programming. When nonlinear coefficient equals 1 and internal friction angle equals 0, the nonlinear Mohr-Coulomb failure criterion degenerates into linear failure criterion. The calculated results of stability number in this work were compared with previous results, and the agreement verifies the effectiveness of the present method. Under the condition of nonlinear Mohr-Coulomb failure criterion, the results show that the supporting force on twin shallow tunnels obviously increases when the nonlinear coefficient, burial depth, ground load or pore water pressure coefficients increase. When the clear distance is 0.5to 1.0 times the diameter of tunnel, the supporting force of twin shallow tunnels reaches its maximum value, which means that the tunnels are the easiest to collapse. While the clear distance increases to 3.5 times the diameter of tunnel, the calculation for twin shallow tunnels can be carried out by the method for independent single shallow tunnel. Therefore, 3.5 times the diameter of tunnel serves as a critical value to determine whether twin shallow tunnels influence each other. In designing twin shallow tunnels,appropriate clear distance value must be selected according to its change rules and actual topographic conditions, meanwhile, the influences of nonlinear failure criterion of soil materials and pore water must be completely considered. During the excavation process, supporting system should be intensified at the positions of larger burial depth or ground load to avoid collapses.
基金Supported by the Sichuan Youth Science ＆ Technology Foundation （2011JTD0009） the National Natural Science Foundation of China （51074138）
文摘Hydraulic fracturing is designed to form a high-conductivity fracture. The proppant will embed into the formation rock, especially in soft rock, owing to the interaction between proppant and fracture surface after fracture closure. Proppant embedment would reduce the fracture width and then lower the fracture conductivity. According to dimensional analysis, the rock is assumed to be an elastic material. Using the theory of elasticity to describe the stage of elastic deformation and analysis of the corresponding simplified embedding process, the study establish the static computation model of the two-dimensional infinite half plane and three-dimensional infinite half space model of the proppant embedment. According to laboratory results, the calculation model was modified, got an effective correction factor and analyzed the causes of errors, then discussed the factors which have impact on proppant embedment. The result calculated by the model in this paper can be reference of prop- pant optimization in on-site fracturing for a certainty degree.
