Coral reef limestone at different depositional depths and facies differ remarkably on the textural and mineralogical characteristics,owing to the complex sedimentary diagenesis.To explore the effects of pore structure...Coral reef limestone at different depositional depths and facies differ remarkably on the textural and mineralogical characteristics,owing to the complex sedimentary diagenesis.To explore the effects of pore structure and mineral composition associated with diagenetic variation on the mechanical behavior of reef limestone,a series of quasi-static and dynamic compression tests along with microscopic examinations were performed on the reef limestone at shallow and deep burial depths.It is revealed that the shallow reef limestone(SRL)is classified as a porous aragonite-type carbonate rock with high porosity(55.3±3.2)%and pore connectivity.In comparison,the deep reef limestone(DRL)is mainly composed of dense calcite-type calcium carbonate with low porosity(4.9±1.6)%and pore connectivity.The DRL strengthened and stiffened by the tight grain framework consistently displays much higher values of the dynamic compressive strength,elastic modulus,brittleness index,and specific energy absorption than those of the SRL.The gap between two types of limestone further increases with an increase in strain rate.It appears that the failure pattern of SRL is dominated by the inherent defects like weak bonding interfaces and growth lines,revealed by the intricate fracturing network and mixed failure.Likewise,although the preexisting megapores in DRL may affect the crack propagation on pore tips to a certain distance,it hardly alters the axial splitting failure of DRL under impacts.The stress wave propagation and attenuation in SRL is primarily controlled by the reflection and diffusion caused by plenty mesopores,as well as an energy dissipation in layer-wise pore collapse and adjacent grain crushing,while the stress wave in DRL is highly hinged on the insulation and diffraction induced by the isolated megapores.This process is accompanied by the energy dissipation behavior of inelastic deformation resulted from the pore-emanated microcracking.展开更多
Two-dimensional(2D)WSe_(2)has received increasing attention due to its unique optical properties and bipolar behavior.Several WSe_(2)-based heterojunctions exhibit bidirectional rectification characteristics,but most ...Two-dimensional(2D)WSe_(2)has received increasing attention due to its unique optical properties and bipolar behavior.Several WSe_(2)-based heterojunctions exhibit bidirectional rectification characteristics,but most devices have a lower rectification ratio.In this work,the Bi_(2)O_(2)Se/WSe_(2)heterojunction prepared by us has a typeⅡband alignment,which can vastly suppress the channel current through the interface barrier so that the Bi_(2)O_(2)Se/WSe_(2)heterojunction device has a large rectification ratio of about 10^(5).Meanwhile,under different gate voltage modulation,the current on/off ratio of the device changes by nearly five orders of magnitude,and the maximum current on/off ratio is expected to be achieved 106.The photocurrent measurement reveals the behavior of recombination and space charge confinement,further verifying the bidirectional rectification behavior of heterojunctions,and it also exhibits excellent performance in light response.In the future,Bi_(2)O_(2)Se/WSe_(2)heterojunction field-effect transistors have great potential to reduce the volume of integrated circuits as a bidirectional controlled switching device.展开更多
在自动驾驶场景下的3D目标检测任务中,探索毫米波雷达数据作为RGB图像输入的补充正成为多模态融合的新兴趋势。然而,现有的毫米波雷达-相机融合方法高度依赖于相机的一阶段检测结果,导致整体性能不够理想。本文提供了一种不依赖于相机...在自动驾驶场景下的3D目标检测任务中,探索毫米波雷达数据作为RGB图像输入的补充正成为多模态融合的新兴趋势。然而,现有的毫米波雷达-相机融合方法高度依赖于相机的一阶段检测结果,导致整体性能不够理想。本文提供了一种不依赖于相机检测结果的鸟瞰图下双向融合方法(BEV-radar)。对于来自不同域的两个模态的特征,BEV-radar设计了一个双向的基于注意力的融合策略。具体地,以基于BEV的3D目标检测方法为基础,我们的方法使用双向转换器嵌入来自两种模态的信息,并根据后续的卷积块强制执行局部空间关系。嵌入特征后,BEV特征在3D对象预测头中解码。我们在nu Scenes数据集上评估了我们的方法,实现了48.2 m AP和57.6 NDS。结果显示,与仅使用相机的基础模型相比,不仅在精度上有所提升,特别地,速度预测误差项有了相当大的改进。代码开源于https://github.com/Etah0409/BEV-Radar。展开更多
基金supported by the National Natural Science Foundation for Excellent Young Scholars of China(No.52222110)the Natural Science Foundation of Jiangsu Province(No.BK20211230).
文摘Coral reef limestone at different depositional depths and facies differ remarkably on the textural and mineralogical characteristics,owing to the complex sedimentary diagenesis.To explore the effects of pore structure and mineral composition associated with diagenetic variation on the mechanical behavior of reef limestone,a series of quasi-static and dynamic compression tests along with microscopic examinations were performed on the reef limestone at shallow and deep burial depths.It is revealed that the shallow reef limestone(SRL)is classified as a porous aragonite-type carbonate rock with high porosity(55.3±3.2)%and pore connectivity.In comparison,the deep reef limestone(DRL)is mainly composed of dense calcite-type calcium carbonate with low porosity(4.9±1.6)%and pore connectivity.The DRL strengthened and stiffened by the tight grain framework consistently displays much higher values of the dynamic compressive strength,elastic modulus,brittleness index,and specific energy absorption than those of the SRL.The gap between two types of limestone further increases with an increase in strain rate.It appears that the failure pattern of SRL is dominated by the inherent defects like weak bonding interfaces and growth lines,revealed by the intricate fracturing network and mixed failure.Likewise,although the preexisting megapores in DRL may affect the crack propagation on pore tips to a certain distance,it hardly alters the axial splitting failure of DRL under impacts.The stress wave propagation and attenuation in SRL is primarily controlled by the reflection and diffusion caused by plenty mesopores,as well as an energy dissipation in layer-wise pore collapse and adjacent grain crushing,while the stress wave in DRL is highly hinged on the insulation and diffraction induced by the isolated megapores.This process is accompanied by the energy dissipation behavior of inelastic deformation resulted from the pore-emanated microcracking.
基金This work was supported by the National Natural Science Foundation of China(61704054,92161115,62374099,and 62022047)the Fundamental Research Funds for the Central Universities(JB2020MS042 and JB2019MS051).
文摘Two-dimensional(2D)WSe_(2)has received increasing attention due to its unique optical properties and bipolar behavior.Several WSe_(2)-based heterojunctions exhibit bidirectional rectification characteristics,but most devices have a lower rectification ratio.In this work,the Bi_(2)O_(2)Se/WSe_(2)heterojunction prepared by us has a typeⅡband alignment,which can vastly suppress the channel current through the interface barrier so that the Bi_(2)O_(2)Se/WSe_(2)heterojunction device has a large rectification ratio of about 10^(5).Meanwhile,under different gate voltage modulation,the current on/off ratio of the device changes by nearly five orders of magnitude,and the maximum current on/off ratio is expected to be achieved 106.The photocurrent measurement reveals the behavior of recombination and space charge confinement,further verifying the bidirectional rectification behavior of heterojunctions,and it also exhibits excellent performance in light response.In the future,Bi_(2)O_(2)Se/WSe_(2)heterojunction field-effect transistors have great potential to reduce the volume of integrated circuits as a bidirectional controlled switching device.
文摘在自动驾驶场景下的3D目标检测任务中,探索毫米波雷达数据作为RGB图像输入的补充正成为多模态融合的新兴趋势。然而,现有的毫米波雷达-相机融合方法高度依赖于相机的一阶段检测结果,导致整体性能不够理想。本文提供了一种不依赖于相机检测结果的鸟瞰图下双向融合方法(BEV-radar)。对于来自不同域的两个模态的特征,BEV-radar设计了一个双向的基于注意力的融合策略。具体地,以基于BEV的3D目标检测方法为基础,我们的方法使用双向转换器嵌入来自两种模态的信息,并根据后续的卷积块强制执行局部空间关系。嵌入特征后,BEV特征在3D对象预测头中解码。我们在nu Scenes数据集上评估了我们的方法,实现了48.2 m AP和57.6 NDS。结果显示,与仅使用相机的基础模型相比,不仅在精度上有所提升,特别地,速度预测误差项有了相当大的改进。代码开源于https://github.com/Etah0409/BEV-Radar。