We study theoretically the electrical shot noise properties of tunnel junctions between a normal metal and a superconductor with the mixture of singlet s-wave and chiral triplet p-wave pairing due to broken inversion ...We study theoretically the electrical shot noise properties of tunnel junctions between a normal metal and a superconductor with the mixture of singlet s-wave and chiral triplet p-wave pairing due to broken inversion symmetry. We investigate how the shot noise properties vary as the relative amplitude between the two parity components in the pairing potential is changed. It is demonstrated that some characteristics of the electrical shot noise properties of such tunnel junctions may depend sensitively on the relative amplitude between the two parity components in the pairing potential, and some significant changes may occur in the electrical shot noise properties when the relative amplitude between the two parity components is varied from the singlet s-wave pairing dominated regime to the chiral triplet p-wave pairing dominated regime. In the chiral triplet p-wave pairing dominated regime, the ratio of noise power to electric current is close to 2e both in the in-gap and in the out-gap region. In the singlet s-wave pairing dominated regime, the value of this ratio is close to 4e in the inner gap region but may reduce to about 2e in the outer gap region as the relative amplitude of the chiral triplet pairing component is increased. The variations of the differential shot noise with the bias voltage also exhibit some significantly different features in different regimes. Such different features can serve as useful diagnostic tools for the determination of the relative magnitude of the two parity components in the pairing potential.展开更多
To investigate the improvement in the fatigue strength of magnesium alloy by peening methods,magnesium alloy AZ31 was treated by submerged laser peening(SLP),cavitation peening(CP),and shot peening(SP),and the fatigue...To investigate the improvement in the fatigue strength of magnesium alloy by peening methods,magnesium alloy AZ31 was treated by submerged laser peening(SLP),cavitation peening(CP),and shot peening(SP),and the fatigue properties were evaluated by a plane bending fatigue test.In the case of SLP,both the impact induced by laser ablation(LA)and that caused by laser cavitation(LC),which developed after LA,were used.In the present study,the fatigue life at a constant bending stress was examined to determine the suitable coverage.It was found that the fatigue strengths at N=10^(7)for the SLP,CP,and SP specimens treated by each optimum condition were 56%,18%,and 16%higher,respectively,than that of the non-peened(NP)specimen,which was 97 MPa.The key factors in the improvement of fatigue strength by peening methods were work hardening and the introduction of compressive residual stress.展开更多
Taking the real part and the imaginary part of complex sound pressure of the sound field as features,a transfer learning model is constructed.Based on the pre-training of a large amount of underwater acoustic data in ...Taking the real part and the imaginary part of complex sound pressure of the sound field as features,a transfer learning model is constructed.Based on the pre-training of a large amount of underwater acoustic data in the preselected sea area using the convolutional neural network(CNN),the few-shot underwater acoustic data in the test sea area are retrained to study the underwater sound source ranging problem.The S5 voyage data of SWellEX-96 experiment is used to verify the proposed method,realize the range estimation for the shallow source in the experiment,and compare the range estimation performance of the underwater target sound source of four methods:matched field processing(MFP),generalized regression neural network(GRNN),traditional CNN,and transfer learning.Experimental data processing results show that the transfer learning model based on residual CNN can effectively realize range estimation in few-shot scenes,and the estimation performance is remarkably better than that of other methods.展开更多
Now object detection based on deep learning tries different strategies.It uses fewer data training networks to achieve the effect of large dataset training.However,the existing methods usually do not achieve the balan...Now object detection based on deep learning tries different strategies.It uses fewer data training networks to achieve the effect of large dataset training.However,the existing methods usually do not achieve the balance between network parameters and training data.It makes the information provided by a small amount of picture data insufficient to optimize model parameters,resulting in unsatisfactory detection results.To improve the accuracy of few shot object detection,this paper proposes a network based on the transformer and high-resolution feature extraction(THR).High-resolution feature extractionmaintains the resolution representation of the image.Channels and spatial attention are used to make the network focus on features that are more useful to the object.In addition,the recently popular transformer is used to fuse the features of the existing object.This compensates for the previous network failure by making full use of existing object features.Experiments on the Pascal VOC and MS-COCO datasets prove that the THR network has achieved better results than previous mainstream few shot object detection.展开更多
Shot peening is a surface modification technology with the metal surface nano machine(SNC),which can modify the surface microstructure and extend the fatigue life of Cu-19Ni alloy.The hardness,damage evolution and mec...Shot peening is a surface modification technology with the metal surface nano machine(SNC),which can modify the surface microstructure and extend the fatigue life of Cu-19Ni alloy.The hardness,damage evolution and mechanical properties were investigated and characterized by scanning electron microscope(SEM),laser confocal microscope(LSM)and material surface performance tester(CFT).The results showed that the surface roughness and friction coefficient of Cu-19Ni alloy decreased with the increase of shot peening duration and diameter,while the microhardness and strength increased.Moreover,with the increase in shot peening duration and diameter,SEM observation showed that the fracture dimples became smaller,meanwhile,with the increase of small cleavage planes,shear tearing ridges and the thickness of the surface nano layer,the fracture mode gradually evolved from plastic to brittle fracture.The uniaxial tensile test of shot peened Cu-19Ni alloy was carried out by MTS testing machine combined with digital image correlation technology(DIC).The evolution of Cu-19Ni surface damage was analyzed,and the evolution equations describing the damage of large deformation zone and small deformation zone were established.The effect of shot peening on the damage evolution behavior of Cu-19Ni alloy was revealed.展开更多
原型网络直接应用于小样本命名实体识别(few-shot named entity recognition,FEW-NER)时存在以下问题:非实体之间不具有较强的语义关系,对实体和非实体都采用相同的方式构造原型将会造成非实体原型不能准确表示非实体的语义特征;仅使用...原型网络直接应用于小样本命名实体识别(few-shot named entity recognition,FEW-NER)时存在以下问题:非实体之间不具有较强的语义关系,对实体和非实体都采用相同的方式构造原型将会造成非实体原型不能准确表示非实体的语义特征;仅使用平均实体向量表示作为原型的计算方式将难以捕捉语义特征相差较大的同类实体.针对上述问题,提出基于细粒度原型网络的小样本命名实体识别(FEW-NER based on fine-grained prototypical networks,FNFP)方法,有助于提高小样本命名实体识别的标注效果.首先,为不同的查询集样本构造不同的非实体原型,捕捉句子中关键的非实体语义特征,得到更为细粒度的原型,提升模型对非实体的识别效果;然后,设计一个不一致性度量模块以衡量同类实体之间的不一致性,对实体与非实体采用不同的度量函数,从而减小同类样本之间的特征表示,提升原型的特征表示能力;最后,引入维特比解码器捕捉标签转换关系,优化最终的标注序列.实验结果表明,采用基于细粒度原型网络的小样本命名实体识别方法,在大规模小样本命名实体识别数据集FEW-NERD上,较基线方法获得提升;同时在跨领域数据集上验证所提方法在不同领域场景下的泛化能力.展开更多
文摘We study theoretically the electrical shot noise properties of tunnel junctions between a normal metal and a superconductor with the mixture of singlet s-wave and chiral triplet p-wave pairing due to broken inversion symmetry. We investigate how the shot noise properties vary as the relative amplitude between the two parity components in the pairing potential is changed. It is demonstrated that some characteristics of the electrical shot noise properties of such tunnel junctions may depend sensitively on the relative amplitude between the two parity components in the pairing potential, and some significant changes may occur in the electrical shot noise properties when the relative amplitude between the two parity components is varied from the singlet s-wave pairing dominated regime to the chiral triplet p-wave pairing dominated regime. In the chiral triplet p-wave pairing dominated regime, the ratio of noise power to electric current is close to 2e both in the in-gap and in the out-gap region. In the singlet s-wave pairing dominated regime, the value of this ratio is close to 4e in the inner gap region but may reduce to about 2e in the outer gap region as the relative amplitude of the chiral triplet pairing component is increased. The variations of the differential shot noise with the bias voltage also exhibit some significantly different features in different regimes. Such different features can serve as useful diagnostic tools for the determination of the relative magnitude of the two parity components in the pairing potential.
基金This work was partly supported by JSPS KAKENHI,Grant Numbers 20H02021 and 22KK0050.
文摘To investigate the improvement in the fatigue strength of magnesium alloy by peening methods,magnesium alloy AZ31 was treated by submerged laser peening(SLP),cavitation peening(CP),and shot peening(SP),and the fatigue properties were evaluated by a plane bending fatigue test.In the case of SLP,both the impact induced by laser ablation(LA)and that caused by laser cavitation(LC),which developed after LA,were used.In the present study,the fatigue life at a constant bending stress was examined to determine the suitable coverage.It was found that the fatigue strengths at N=10^(7)for the SLP,CP,and SP specimens treated by each optimum condition were 56%,18%,and 16%higher,respectively,than that of the non-peened(NP)specimen,which was 97 MPa.The key factors in the improvement of fatigue strength by peening methods were work hardening and the introduction of compressive residual stress.
基金supported by the National Natural Science Foundation of China(1197428611904274)+1 种基金the Shaanxi Young Science and Technology Star Program(2021KJXX-07)the fundamental research funding for characteristic disciplines(G2022WD0235)。
文摘Taking the real part and the imaginary part of complex sound pressure of the sound field as features,a transfer learning model is constructed.Based on the pre-training of a large amount of underwater acoustic data in the preselected sea area using the convolutional neural network(CNN),the few-shot underwater acoustic data in the test sea area are retrained to study the underwater sound source ranging problem.The S5 voyage data of SWellEX-96 experiment is used to verify the proposed method,realize the range estimation for the shallow source in the experiment,and compare the range estimation performance of the underwater target sound source of four methods:matched field processing(MFP),generalized regression neural network(GRNN),traditional CNN,and transfer learning.Experimental data processing results show that the transfer learning model based on residual CNN can effectively realize range estimation in few-shot scenes,and the estimation performance is remarkably better than that of other methods.
基金the National Natural Science Foundation of China under grant 62172059 and 62072055Hunan Provincial Natural Science Foundations of China under Grant 2020JJ4626+2 种基金Scientific Research Fund of Hunan Provincial Education Department of China under Grant 19B004“Double First-class”International Cooperation and Development Scientific Research Project of Changsha University of Science and Technology under Grant 2018IC25the Young Teacher Growth Plan Project of Changsha University of Science and Technology under Grant 2019QJCZ076.
文摘Now object detection based on deep learning tries different strategies.It uses fewer data training networks to achieve the effect of large dataset training.However,the existing methods usually do not achieve the balance between network parameters and training data.It makes the information provided by a small amount of picture data insufficient to optimize model parameters,resulting in unsatisfactory detection results.To improve the accuracy of few shot object detection,this paper proposes a network based on the transformer and high-resolution feature extraction(THR).High-resolution feature extractionmaintains the resolution representation of the image.Channels and spatial attention are used to make the network focus on features that are more useful to the object.In addition,the recently popular transformer is used to fuse the features of the existing object.This compensates for the previous network failure by making full use of existing object features.Experiments on the Pascal VOC and MS-COCO datasets prove that the THR network has achieved better results than previous mainstream few shot object detection.
基金Funded by Natural Science Foundation of the Inner Mongolia(Nos.2019MS01015,2019MS01017)National Natural Science Foundation of China(No.11002065)。
文摘Shot peening is a surface modification technology with the metal surface nano machine(SNC),which can modify the surface microstructure and extend the fatigue life of Cu-19Ni alloy.The hardness,damage evolution and mechanical properties were investigated and characterized by scanning electron microscope(SEM),laser confocal microscope(LSM)and material surface performance tester(CFT).The results showed that the surface roughness and friction coefficient of Cu-19Ni alloy decreased with the increase of shot peening duration and diameter,while the microhardness and strength increased.Moreover,with the increase in shot peening duration and diameter,SEM observation showed that the fracture dimples became smaller,meanwhile,with the increase of small cleavage planes,shear tearing ridges and the thickness of the surface nano layer,the fracture mode gradually evolved from plastic to brittle fracture.The uniaxial tensile test of shot peened Cu-19Ni alloy was carried out by MTS testing machine combined with digital image correlation technology(DIC).The evolution of Cu-19Ni surface damage was analyzed,and the evolution equations describing the damage of large deformation zone and small deformation zone were established.The effect of shot peening on the damage evolution behavior of Cu-19Ni alloy was revealed.
文摘原型网络直接应用于小样本命名实体识别(few-shot named entity recognition,FEW-NER)时存在以下问题:非实体之间不具有较强的语义关系,对实体和非实体都采用相同的方式构造原型将会造成非实体原型不能准确表示非实体的语义特征;仅使用平均实体向量表示作为原型的计算方式将难以捕捉语义特征相差较大的同类实体.针对上述问题,提出基于细粒度原型网络的小样本命名实体识别(FEW-NER based on fine-grained prototypical networks,FNFP)方法,有助于提高小样本命名实体识别的标注效果.首先,为不同的查询集样本构造不同的非实体原型,捕捉句子中关键的非实体语义特征,得到更为细粒度的原型,提升模型对非实体的识别效果;然后,设计一个不一致性度量模块以衡量同类实体之间的不一致性,对实体与非实体采用不同的度量函数,从而减小同类样本之间的特征表示,提升原型的特征表示能力;最后,引入维特比解码器捕捉标签转换关系,优化最终的标注序列.实验结果表明,采用基于细粒度原型网络的小样本命名实体识别方法,在大规模小样本命名实体识别数据集FEW-NERD上,较基线方法获得提升;同时在跨领域数据集上验证所提方法在不同领域场景下的泛化能力.
