The 3D turbulence k-ε model flow of the steel melt （continuous phase） and the trajectories of individual gas bubbles （dispersed phase） in a continuous casting mold were simulated using an Eulerian-Lagrangian appr...The 3D turbulence k-ε model flow of the steel melt （continuous phase） and the trajectories of individual gas bubbles （dispersed phase） in a continuous casting mold were simulated using an Eulerian-Lagrangian approach. In order to investigate the effect of bubble size distribution, the radii of bubbles are set with an initial value of 0. 1- 2.5 mm which follows the normal distribution. The presented results indicate that, in the submerged entry nozzle （SEN）, the distribution of void fraction is only near the wall. Due to the fact that the bubbles motion is only limited to the wall, the deoxidization products have no access to contacting the wall, which prevents clogging. In the mold, the bubbles with a radius of 0. 25--2.5 mm will move to the top surface. Larger bubbles issuing out of the ports will attack the menis- cus and induce the fluid flows upwards in the top surface near the nozzle. It may induce mold powder entrapment into the mold. The bubbles with a radius of 0.1--0.25 mm will move to the zone near the narrow surface and the wide surface. These small bubbles will probably be trapped by the solidification front. Most of the bubbles moving to the narrow surface will flow with the ascending flow, while others will flow with the descending flow.展开更多
It has been revealed that the different morphologies of anodized TiO_2 nanotubes, especially nanotube diameters, triggered different cell behaviors. However, the influence of TiO_2 nanotubes with coexisting multi-size...It has been revealed that the different morphologies of anodized TiO_2 nanotubes, especially nanotube diameters, triggered different cell behaviors. However, the influence of TiO_2 nanotubes with coexisting multi-size diameters on cell behaviors is seldom reported. In this work, coexisting four-diameter TiO_2 nanotube samples, namely,one single substrate with the integration of four different nanotube diameters(60, 150, 250, and 350 nm), were prepared by repeated anodization. The boundaries between two different diameter regions show well-organized structure without obvious difference in height. The adhesion behaviors of MC3T3-E1 cells on the coexisting fourdiameter TiO_2 nanotube arrays were investigated. The results exhibit a significant difference of cell density between smaller diameters(60 and 150 nm) and larger diameters(250 and 350 nm) within 24 h incubation with the coexistence of different diameters, which is totally different from that on the single-diameter TiO_2 nanotube arrays. The coexistence of four different diameters does not change greatly the cell morphologies compared with the singlediameter nanotubes. The findings in this work are expected to offer further understanding of the interaction between cells and materials.展开更多
In this paper,we propose a new method to achieve automatic matching of multi-scale roads under the constraints of smaller scale data.The matching process is:Firstly,meshes are extracted from two different scales road ...In this paper,we propose a new method to achieve automatic matching of multi-scale roads under the constraints of smaller scale data.The matching process is:Firstly,meshes are extracted from two different scales road data.Secondly,several basic meshes in the larger scale road network will be merged into a composite one which is matched with one mesh in the smaller scale road network,to complete the N∶1(N>1)and 1∶1 matching.Thirdly,meshes of the two different scale road data with M∶N(M>1,N>1)matching relationships will be matched.Finally,roads will be classified into two categories under the constraints of meshes:mesh boundary roads and mesh internal roads,and then matchings between the two scales meshes will be carried out within their own categories according to the matching relationships.The results show that roads of different scales will be more precisely matched using the proposed method.展开更多
从三维Mesh数据中分割建筑物立面以识别对象,是三维场景理解的关键,但现有方法多依赖高成本的精细标注数据。针对该问题,提出了一种半监督学习方法,引入一种基于对比学习和一致性正则化的半监督语义分割(semi-supervised semantic segme...从三维Mesh数据中分割建筑物立面以识别对象,是三维场景理解的关键,但现有方法多依赖高成本的精细标注数据。针对该问题,提出了一种半监督学习方法,引入一种基于对比学习和一致性正则化的半监督语义分割(semi-supervised semantic segmentation based on contrastive learning and consistency regularization,SS_CC)方法,用于分割三维Mesh数据的建筑物立面。在SS_CC方法中,改进后的对比学习模块利用正负样本之间的类可分性,能够更有效地利用类特征信息;提出的基于特征空间的一致性正则化损失函数,从挖掘全局特征的角度增强了对所提取建筑物立面特征的鉴别力。实验结果表明,所提出的SS_CC方法在F1分数、mIoU指标上优于当前一些主流方法,且在建筑物的墙面和窗户上的分割效果相对更好。展开更多
基金Item Sponsored by National Natural Science Foundation of China(50874130,50974034)
文摘The 3D turbulence k-ε model flow of the steel melt （continuous phase） and the trajectories of individual gas bubbles （dispersed phase） in a continuous casting mold were simulated using an Eulerian-Lagrangian approach. In order to investigate the effect of bubble size distribution, the radii of bubbles are set with an initial value of 0. 1- 2.5 mm which follows the normal distribution. The presented results indicate that, in the submerged entry nozzle （SEN）, the distribution of void fraction is only near the wall. Due to the fact that the bubbles motion is only limited to the wall, the deoxidization products have no access to contacting the wall, which prevents clogging. In the mold, the bubbles with a radius of 0. 25--2.5 mm will move to the top surface. Larger bubbles issuing out of the ports will attack the menis- cus and induce the fluid flows upwards in the top surface near the nozzle. It may induce mold powder entrapment into the mold. The bubbles with a radius of 0.1--0.25 mm will move to the zone near the narrow surface and the wide surface. These small bubbles will probably be trapped by the solidification front. Most of the bubbles moving to the narrow surface will flow with the ascending flow, while others will flow with the descending flow.
基金supported by the National Natural Science Foundation of China(No.51401126,No.51271117)Shanghai Committee of Science and Technology,China(No.14441901800)
文摘It has been revealed that the different morphologies of anodized TiO_2 nanotubes, especially nanotube diameters, triggered different cell behaviors. However, the influence of TiO_2 nanotubes with coexisting multi-size diameters on cell behaviors is seldom reported. In this work, coexisting four-diameter TiO_2 nanotube samples, namely,one single substrate with the integration of four different nanotube diameters(60, 150, 250, and 350 nm), were prepared by repeated anodization. The boundaries between two different diameter regions show well-organized structure without obvious difference in height. The adhesion behaviors of MC3T3-E1 cells on the coexisting fourdiameter TiO_2 nanotube arrays were investigated. The results exhibit a significant difference of cell density between smaller diameters(60 and 150 nm) and larger diameters(250 and 350 nm) within 24 h incubation with the coexistence of different diameters, which is totally different from that on the single-diameter TiO_2 nanotube arrays. The coexistence of four different diameters does not change greatly the cell morphologies compared with the singlediameter nanotubes. The findings in this work are expected to offer further understanding of the interaction between cells and materials.
基金The National Natural Science Foundation of China(Nos.4110136241471386)。
文摘In this paper,we propose a new method to achieve automatic matching of multi-scale roads under the constraints of smaller scale data.The matching process is:Firstly,meshes are extracted from two different scales road data.Secondly,several basic meshes in the larger scale road network will be merged into a composite one which is matched with one mesh in the smaller scale road network,to complete the N∶1(N>1)and 1∶1 matching.Thirdly,meshes of the two different scale road data with M∶N(M>1,N>1)matching relationships will be matched.Finally,roads will be classified into two categories under the constraints of meshes:mesh boundary roads and mesh internal roads,and then matchings between the two scales meshes will be carried out within their own categories according to the matching relationships.The results show that roads of different scales will be more precisely matched using the proposed method.
文摘从三维Mesh数据中分割建筑物立面以识别对象,是三维场景理解的关键,但现有方法多依赖高成本的精细标注数据。针对该问题,提出了一种半监督学习方法,引入一种基于对比学习和一致性正则化的半监督语义分割(semi-supervised semantic segmentation based on contrastive learning and consistency regularization,SS_CC)方法,用于分割三维Mesh数据的建筑物立面。在SS_CC方法中,改进后的对比学习模块利用正负样本之间的类可分性,能够更有效地利用类特征信息;提出的基于特征空间的一致性正则化损失函数,从挖掘全局特征的角度增强了对所提取建筑物立面特征的鉴别力。实验结果表明,所提出的SS_CC方法在F1分数、mIoU指标上优于当前一些主流方法,且在建筑物的墙面和窗户上的分割效果相对更好。
