检索结果-南通市图书馆

Semi-supervised remote sensing image scene classification with prototype-based consistency

在线全文

学校读者我要写书评

暂无评论

Chinese Journal of Aeronautics 2024年第2期37卷 459-470页

作者： Yang LI Zhang LI zi wang Kun wang Qifeng YU College of Aerospace Science and Engineering National University of Defense TechnologyChangsha 410073China

Deep learning significantly improves the accuracy of remote sensing image scene classification,benefiting from the large-scale ***,annotating the remote sensing images is time-consuming and even tough for *** neural networks trained using a few labeled samples usually generalize less to new unseen *** this paper,we propose a semi-supervised approach for remote sensing image scene classification based on the prototype-based consistency,by exploring massive unlabeled *** this end,we,first,propose a feature enhancemen. module to extract discriminative *** is achieved by focusing the model on the foreground ***,the prototype-based classifier is introduced to the framework,which is used to acquire consistent feature *** conduct a series of experimen.s on NWPU-RESISC45 and Aerial Image Dataset(AID).Our method improves the State-Of-The-Art(SOTA)method on NWPU-RESISC45 from 92.03%to 93.08%and on AID from 94.25%to 95.24%in terms of accuracy.

关键词： Semi-supervised learning Remote sensing Scene classification Prototype network Deep learning

Rapid prediction of damaged ship roll motion responses in beam waves based on stacking algorithm

在线全文

学校读者我要写书评

暂无评论

Journal of Hydrodynamics 2024年第2期36卷 394-405页

作者： Xin-ran Liu Ting-qiu Li zi-ping wang School of Naval Architecture Ocean and Energy Power EngineeringWuhan University of TechnologyWuhan430063China

Accurate modeling for highly non-linear coupling of a damaged ship with liquid sloshing in waves is still of considerable interest within the computational fluid dynamics(CFD)and AI *** paper describes a data-driven Stacking algorithm for fast prediction of roll motion response amplitudes in beam waves by constructing a hydrodynamics model of a damaged ship based on the dynamic overlapping grid CFD *** general idea is to optimize various parameters varying with four types of classical base models like multi-layer perception,support vector regression,random forest,and hist gradient boosting *** offers several attractive properties in terms of accuracy and efficiency by choosing the standard DTMB 5415 model with double damaged compartmen.s for *** is clearly demonstrated that the predicted response amplitude operator(RAO)in the regular beam waves agrees well with the experimen.al data available,which verifies the accuracy of the established damaged ship hydrodynamics *** high-quality CFD samples,therefore,implemen.ation of the designed Stacking algorithm with its optimal combination can predict the damaged ship roll motion amplitudes effectively and accurately(e.g.,the coefficient of determination 0.9926,the average absolute error 0.0955 and CPU 3s),by comparison of four types of typical base models and their various ***,the established Stacking algorithm provides one potential that can break through problems involving the time-consuming and low efficiency for large-scale lengthy CFD simulations.

关键词： Ship motion damaged ship computational fluid dynamics(CFD) machine learning stacking algorithm

Thermodynamic and experimen.al study of high-temperature roasting of blast furnace gas ash for recovery of metallic zinc and iron

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Iron and Steel Research International 2024年第1期31卷 108-120页

作者： zi-han wang Jing Guo Han-jie Guo Bin Li School of Metallurgical Engineering University of Science and Technology BeijingBeijing100083China School of Metallurgical Engineering Xi’an University of Architecture and TechnologyXi’an710055ShaanxiChina

A high-temperature reduction roasting method was used to achieve metallic iron and zinc recovery from blast furnace gas ash(BFA).The reduction processes for Zn-containing and Fe-containing oxides were analyzed in detail by using ther-modynamic equilibrium calculation and the principle of minimum free *** results showed that the main reaction in the system is the reduction of ZnFe_(2)_(4)and iron *** the full temperature range,iron oxides were more easily reduced than zinc *** of the amount of CO contained in the system,the reduction of ZnO to Zn was difficult to proceed below the boiling point(906℃)of *** the reduction temperature is below 906℃,the reduction process of zinc ferrate was ZnFe_(2)_(4)→ZnO;when the reduction temperature is above 906℃,its reduction process becomed ZnFe_(2)_(4)→ZnO→Zn(g).The metallization and dezincification rates of the BFA gradually increased with increasing reaction *** the C/O ratio increased,the metallization and dezincification rates first increased and then *** effect of reduction time on BFA reduction was similar to that of reaction temperature.

关键词： Blast furnace gas ash Thermodynamics zinc removal Iron reduction High-temperature reduction roasting Integrated utilization

Reliable calculations of nuclear binding energies by the Gaussian process of machine learning

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Nuclear Science and Techniques 2024年第6期35卷 130-144页

作者： zi-Yi Yuan Dong Bai Zhen wang Zhong-Zhou Ren School of Physics Science and Engineering Tongji UniversityShanghai 200092China College of Science Hohai UniversityNanjing 211100China Key Laboratory of Advanced Micro-Structure Materials Ministry of EducationShanghai 200092China

Reliable calculations of nuclear binding energies are crucial for advancing the research of nuclear physics. Machine learning provides an innovative approach to exploring complex physical problems. In this study, the nuclear binding energies are modeled directly using a machine-learning method called the Gaussian process. First, the binding energies for 2238 nuclei with Z > 20 and N > 20 are calculated using the Gaussian process in a physically motivated feature space, yielding an average deviation of 0.046 MeV and a standard deviation of 0.066 MeV. The results show the good learning ability of the Gaussian process in the studies of binding energies. Then, the predictive power of the Gaussian process is studied by calculating the binding energies for 108 nuclei newly included in AME2020. The theoretical results are in good agreemen. with the experimen.al data, reflecting the good predictive power of the Gaussian process. Moreover, the α-decay energies for 1169 nuclei with 50 ≤ Z ≤ 110 are derived from the theoretical binding energies calculated using the Gaussian process. The average deviation and the standard deviation are, respectively, 0.047 MeV and 0.070 MeV. Noticeably, the calculated α-decay energies for the two new isotopes ^ (204 )Ac(Huang et al. Phys Lett B 834, 137484(2022)) and ^ (207) Th(Yang et al. Phys Rev C 105, L051302(2022)) agree well with the latest experimen.al data. These results demonstrate that the Gaussian process is reliable for the calculations of nuclear binding energies. Finally, the α-decay properties of some unknown actinide nuclei are predicted using the Gaussian process. The predicted results can be useful guides for future research on binding energies and α-decay properties.

关键词： Nuclear binding energies decay Machine learning Gaussian process

Microstructure and mechanical properties of a cast TRIP-assisted multiphase stainless steel

在线全文

学校读者我要写书评

暂无评论

China Foundry 2024年第3期21卷 221-228页

作者： men.-xin wang zi-xiang Wu Jing-yu He Xiang Chen School of Materials Science and Engineering Tsinghua UniversityBeijing 100084China Key Laboratory for Advanced Materials Processing Technology Ministry of EducationBeijing 100084China

Stainless steels are used in a wide range of complex environmen.s due to their excellent corrosion *** stainless steels can offer an excellent combination of strength,toughness and corrosion resistance due to the coexistence of different *** microstructure and mechanical properties of a novel cast multiphase stainless steel,composed of martensite,ferrite,and austenite,were investigated following appropriate heat treatmen. processes:solution treatmen. at 1,050℃ for 0.5 h followed by water quenching to room temperature,and aging treatmen. at 500℃ for 4 h followed by water quenching to room *** show reversed austenite is formed by diffusion of Ni elemen. during aging process,and the enrichmen. of Ni atoms directly determines the mechanical stability of *** austenite with a lower Ni content undergoes a martensitic transformation during plastic *** tensile strength of the specimen.exceeds 1,100 MPa and the elongation exceeds 24%after solid solution,and further increases to 1,247 MPa and 25%after aging *** enhancemen. is due to the TRIP effect of austenite and the precipitation of the nanoscale G-phase pinning dislocations in ferrite and martensite.

关键词： multiphase stainless steel mechanical properties TRIP effect reversed austenite G-phase

A Method of Generating Semi-Experimen.al Biomedical Datasets

在线全文

学校读者我要写书评

暂无评论

Journal of Beijing Institute of Technology 2024年第3期33卷 219-226页

作者： Jing wang Naike Du zi He Xiuzhu Ye School of Information and Electronics Beijing Institute of TechnologyBeijing 100081China

This paper proposed a method to generate semi-experimen.al biomedical datasets based on full-wave simulation *** system noise such as antenna port couplings is fully considered in the proposed datasets,which is more realistic than synthetical *** this paper,datasets containing different shapes are constructed based on the relative permittivities of human ***,a back-propagation scheme is used to obtain the rough reconstructions,which will be fed into a U-net convolutional neural network(CNN)to recover the high-resolution *** results show that the network trained on the datasets generated by the proposed method can obtain satisfying reconstruction results and is promising to be applied in real-time biomedical imaging.

关键词： electromagnetic imaging dataset biomedical imaging

Cavitation research with computational fluid dynamics:From Euler-Euler to Euler-Lagrange approach

在线全文

学校读者我要写书评

暂无评论

Journal of Hydrodynamics 2024年第1期36卷 1-23页

作者： Bin Ji zi-yang wang Huai-yu Cheng Rickard E.Bensow State Key Laboratory of Water Resources Engineering and Managemen. Wuhan UniversityWuhan430072China Departmen. of Mechanics and Maritime Sciences Chalmers University of TechnologyGothenburgSweden

Unsteady cavitating flow often contains vapor structures with a wide range of different length scales,from micro-bubbles to large cavities,which issues a big challenge to precisely investigate its evolution mechanism by computational fluid dynamics(CFD)*** present work reviews the developmen. of simulation methods for cavitation,especially the emerging Euler-Lagrange ***,the progress of the numerical investigation of hot and vital issues is discussed,including cavitation inception,cloud cavitation inner structure and its formation mechanism,cavitation erosion,and cavitation *** is indicated that the Euler-Lagrange method can determine cavitation inception point *** cloud cavitation,the Euler-Lagrange method can reveal the source of microbubbles and their distribution law inside the shedding *** method also has advantages and great potential in assessing cloud cavitation-induced erosion and *** the ever-growing demands of cavitation simulation accuracy in basic research and engineering applications,how to improve the Euler-Lagrange method’s stability and applicability is still an open *** further promote the application of this advanced CFD simulation technology in cavitation research,some key issues are to be solved and feasible suggestions are put forward for further work.

关键词： numerical simulation Euler-Euler method Euler-Lagrange method bubble dynamicsgy,Gothenburg,Sweden

Exploring the Pathogenesis of Autoimmune Liver Diseases from the Heterogeneity of Target Cells

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Clinical and Translational Hepatology 2024年第7期12卷 659-666页

作者： zi-Xuan Qiu Lin-Xiang Huang Xiao-Xiao wang zi-Long wang Xiao-He Li Bo Feng Peking University People’s Hospital Peking University Hepatology InstituteBeijing Key Laboratory of Hepatitis C and Immunotherapy for Liver DiseasesBeijing International Cooperation Base for Science and Technology on NAFLD DiagnosisBeijingChina

The incidence of autoimmune liver diseases(ALDs)and research on their pathogenesis are increasing ***,except for autoimmune hepatitis,which responds well to immunosuppression,primary biliary cholangitis and primary sclerosing cholangitis are insensitive to immunosuppressive *** the known effects of the environmen.,genetics,and immunity on ALDs,the heterogeneity of target cells provides new insights into their *** review started by exploring the heterogeneity in the developmen.,structures,and functions of hepatocytes and epithelial cells of the small and large bile *** example,cytokeratin(CK)8 and CK18 are primarily expressed in hepatocytes,while CK7 and CK19 are primarily expressed in intrahepatic ***,emerging technologies of single-cell RNA sequencing and spatial transcriptomic are being applied to study *** review offered a new perspective on understanding the pathogenic mechanisms and potential treatmen. strategies for ALDs.

关键词： Autoimmune hepatitis Primary biliary cholangitis Primary sclerosing cholangitis Disease heterogeneity Single-cell RNA sequencing Pathogenesis

Extrusion 3D printing of carbon nanotube-assembled carbon aerogel nanocomposites with high electrical conductivity

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Nano Materials Science 2024年第3期6卷 312-319页

作者： Lukai wang Jing men.Junzong Feng Yonggang Jiang Liangjun Li Yijie Hu Jian Feng Science and Technology on Advanced Ceramic Fibers and Composites Laboratory College of Aerospace Science and EngineeringNational University of Defense Technology109 De Ya RdChangshaHunan410073PR China

Carbon nanotubes(CNTs)with high aspect ratio and excellent electrical conduction offer huge functional improvemen.s for current carbon ***,there remains a major challenge for achieving the on-demand shaping of carbon aerogels with tailored micro-nano structural textures and geometric ***,a facile extrusion 3D printing strategy has been proposed for fabricating CNT-assembled carbon(CNT/C)aerogel nanocomposites through the extrusion printing of pseudoplastic carbomer-based inks,in which the stable dispersion of CNT nanofibers has been achieved relying on the high viscosity of carbomer *** extrusion printing,the chemical solidification through polymerizing RF sols enables 3D-printed aerogel nanocomposites to display high shape fidelity in macroscopic *** from the micro-nano scale assembly of CNT nanofiber networks and carbon nanoparticle networks in composite phases,3D-printed CNT/C aerogels exhibit enhanced mechanical strength(fracture strength,0.79 MPa)and typical porous structure characteristics,including low density(0.220 g cm^(-3)),high surface area(298.4 m^(2)g^(-1)),and concentrated pore diameter distribution(~32.8nm).More importantly,CNT nanofibers provide an efficient electron transport pathway,imparting 3D-printed CNT/C aerogel composites with a high electrical conductivity of 1.49 S cm^(-1).Our work would offer feasible guidelines for the design and fabrication of shape-dominated functional materials by additive manufacturing.

关键词： Carbon aerogel Extrusion 3D printing Carbon nanotube Electrical conductivity Rheology