检索结果-南通市图书馆

维普期刊数据库

Communication efficiency optimization of federated learning for computing and network convergence of 6G networks

在线全文

学校读者我要写书评

暂无评论

Frontiers of Information Technology & Electronic Engineering 2024年第5期25卷 713-727页

作者： Yizhuo CAI Bo LEI Qianying ZHAO Jing PENG min WEI Yushun zhang xing zhang Wireless Signal Processing and Network Laboratory Beijing University of Posts and TelecommunicationsBeijing 100876China Research Institute of China Telecom Co. Ltd.Beijing 102209China Beijing Telecom Beijing 100011China

Federated learning effectively addresses issues such as data privacy by collaborating across participating devices to train global ***,factors such as network topology and computing power of devices can affect its training or communication process in complex network *** and network convergence(CNC)of sixth-generation(6G)networks,a new network architecture and paradigm with computing-measurable,perceptible,distributable,dispatchable,and manageable capabilities,can effectively support federated learning training and improve its communication *** guiding the participating devices'training in federated learning based on business requirements,resource load,network conditions,and computing power of devices,CNC can reach this *** this paper,to improve the communication eficiency of federated learning in complex networks,we study the communication eficiency optimization methods of federated learning for CNC of 6G networks that give decisions on the training process for different network conditions and computing power of participating *** simulations address two architectures that exist for devices in federated learning and arrange devices to participate in training based on arithmetic power while achieving optimization of communication efficiency in the process of transferring model *** results show that the methods we proposed can cope well with complex network situations,effectively balance the delay distribution of participating devices for local training,improve the communication eficiency during the transfer of model parameters,and improve the resource utilization in the network.

关键词： Computing and network convergence Communication efficiency Federated learning Two architectures

CrossFormer Embedding DeepLabv3+ for Remote Sensing Images Semantic Segmentation

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2024年第4期79卷 1353-1375页

作者： Qixiang Tong Zhipeng Zhu min zhang Kerui Cao Haihua xing School of Information Science and Technology Hainan Normal UniversityHaikou571158China

High-resolution remote sensing image segmentation is a challenging task. In urban remote sensing, the presenceof occlusions and shadows often results in blurred or invisible object boundaries, thereby increasing the difficultyof segmentation. In this paper, an improved network with a cross-region self-attention mechanism for multi-scalefeatures based onDeepLabv3+is designed to address the difficulties of small object segmentation and blurred targetedge segmentation. First,we use CrossFormer as the backbone feature extraction network to achieve the interactionbetween large- and small-scale features, and establish self-attention associations between features at both large andsmall scales to capture global contextual feature information. Next, an improved atrous spatial pyramid poolingmodule is introduced to establish multi-scale feature maps with large- and small-scale feature associations, andattention vectors are added in the channel direction to enable adaptive adjustment of multi-scale channel *** proposed networkmodel is validated using the PotsdamandVaihingen datasets. The experimental results showthat, compared with existing techniques, the network model designed in this paper can extract and fuse multiscaleinformation, more clearly extract edge information and small-scale information, and segment boundariesmore smoothly. Experimental results on public datasets demonstrate the superiority of ourmethod compared withseveral state-of-the-art networks.

关键词： Semantic segmentation remote sensing multiscale self-attention

Electronic structure and spin state regulation of vanadium nitride via a sulfur doping strategy toward flexible zinc-air batteries

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 Energy Chemistry 2024年第2期89卷 239-249,I0007页

作者： Daijie Deng Honghui zhang Jianchun Wu xing Tang min Ling Sihua Dong Li Xu Henan Li Huaming Li Institute for Energy Research Institute of Quantum and Sustainable TechnologySchool of Chemistry and Chemical EngineeringSchool of Materials Science and EngineeringJiangsu Province and Education Ministry Co-Sponsored Synergistic Innovation Center of Modern Agricultural EquipmentJiangsu UniversityZhenjiang 212013JiangsuChina Key Laboratory of Radio Frequency and Micro-Nano Electronics of Jiangsu Province Nanjing 210023JiangsuChina Zhenjiang Kangyuan Electrical Appliance Co. Ltd.Zhenjiang 212013JiangsuChina College of Chemical and Biological Engineering Zhejiang UniversityHangzhou 310027ZhejiangChina YTO Group Corporation Dongfanghong(Henan)Agricultural Service Technology Co. Ltd.Luoyang 471033HenanChina

Owing to the distinctive structural characteristics,vanadium nitride(VN)is highly regarded as a catalyst for oxygen reduction reaction(ORR)in zinc-air batteries(ZABs).However,VN exhibits limited intrinsic ORR activity due to the weak adsorption ability to O-containing ***,the S-doped VN anchored on N,S-doped multi-dimensional carbon(S-VN/Co/NS-MC)was constructed using the solvothermal and in-situ doping *** sulfur atoms into VN species alters the electron spin state of vanadium in the S-VN/Co/NS-MC for regulating the adsorption energy of vanadium sites to oxygen *** introduced sulfur atoms polarize the V 3d_(z)^(2) electrons,shifting spin-down electrons closer to the Fermi level in the S-VN/Co/***,the introduction of sulfur atoms into VN species enhances the adsorption energy of vanadium sites for oxygen ****OOH dissociation transitions from being unspontaneous on the VN surface to a spontaneous state on the S-doped VN ***,the ORR barrier on the S-VN/Co/NS-MC surface is *** S-VN/Co/NS-MC demonstrates a higher half-wave potential and limiting current density compared to the VN/Co/*** S-VN/Co/NS-MC-based liquid ZABs display a power density of 195.7 m W cm^(-2),a specific capacity of 815.7 m A h g^(-1),and a cycling stability exceeding 250 *** S-VN/Co/NS-MC-based flexible ZABs are successfully employed to charge both a smart watch and a mobile *** approach holds promise for advancing the commercial utilization of VN-based catalysts in ZABs.

关键词： S-doped VN Electronic structures Spin state regulation Oxygen reduction reaction Zinc-air batteries

维普期刊数据库

An optimization-oriented modeling approach using input convex neural networks and its application on optimal chiller loading

在线全文

学校读者我要写书评

暂无评论

Building Simulation 2024年第4期17卷 639-655页

作者： Shanshuo xing Jili zhang Song Mu Dalian University of Technology Dalian116024China Guangdong Airport Baiyun Information Technology Co. Ltd.Guangzhou510000China

Optimization for the multi-chiller system is an indispensable approach for the operation of highly efficient chiller *** optima obtained by model-based optimization algorithms are dependent on precise and solvable objective *** classical neural networks cannot provide convex input-output mappings despite capturing impressive nonlinear fitting capabilities,resulting in a reduction in the robustness of model-based *** this paper,we leverage the input convex neural networks(ICNN)to identify the chiller model to construct a convex mapping between control variables and the objective function,which enables the NN-based OCL as a convex optimization problem and apply it to multi-chiller optimization for optimal chiller loading(OCL).Approximation performances are evaluated through a four-model comparison based on an experimental data set,and the statistical results show that,on the premise of retaining prior convexities,the proposed model depicts excellent approximation power for the data set,especially the unseen ***,the ICNN model is applied to a typical OCL problem for a multi-chiller system and combined with three types of optimization *** with conventional and meta-heuristic methods,the numerical results suggest that the gradient-based BFGS algorithm provides better energy-saving ratios facing consecutive cooling load inputs and an impressive convergence speed.

关键词： chiller plant input convex neural network optimal load distribution convex optimization

Numerical simulation study on the hygrothermal performance of building exterior walls under dynamic wind-driven rain condition

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Building Simulation 2024年第2期17卷 207-221页

作者： xing Hu Huibo zhang Hui Yu School of Design Shanghai Jiao Tong UniversityShanghai200240China

Wind-driven rain(WDR)has a significant influence on the hygrothermal performance,durability,and energy consumption of building *** calculation of WDR loads using semi-empirical models has been incorporated into the boundary conditions of coupled heat and moisture transfer ***,prior research often relied on fixed WDR absorption ratio,which fail to accurately capture the water absorption characteristics of porous building materials under rainfall ***,this study aims to investigate the coupled heat and moisture transfer of exterior walls under dynamic WDR boundary conditions,utilizing an empirically obtained WDR absorption ratio model based on field *** developed coupled heat and moisture transfer model is validated against the HAMSTAD *** findings reveal that the total WDR flux calculated with the dynamic WDR boundary is lower than that obtained with the fixed WDR boundary,with greater disparities observed in orientations experiencing higher WDR *** variations in moisture flow significantly impact the surface temperature and relative humidity of the walls,influencing the calculation of cooling and heating loads by different *** to the transient heat transfer model,the coupled heat and moisture transfer model incorporating dynamic WDR boundary exhibits maximum increases of 17.6%and 16.2%in cooling and heating loads,*** dynamic WDR boundary conditions provide more precise numerical values for surface moisture flux,offering valuable insights for the thermal design of building enclosures and load calculations for HVAC systems.

关键词： wind-driven rain building component hygrothermal model transient simulation cooling and heating loads

A2B7-type La-Mg-Ni alloys prepared by Mg thermal diffusion for improved hydrogen storage performance

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Rare Metals 2024年第7期43卷 3260-3272页

作者： Yong-Xi zhang Guan-Jiu Wu Jing Gu Hong-xing Kang Yuan Li Di Zhou Wen-Feng Wang Lu zhang Shu-min Han Hebei Key Laboratory of Applied Chemistry School of Environmental and Chemical EngineeringYanshan University Department of Chemistry and Biochemistry San Diego State University State Key Laboratory of Metastable Materials Science and Technology Yanshan University

A novel approach based on thermal diffusion was used to achieve controllable Mg content in A2B7-type La-Mg-Ni-based *** formation mechanism of the A2B7-type phase as a result of the thermal diffusion process and the effect of Mg content on hydrogen storage performance were investigated.X-ray diffraction（XRD） patterns and Rietveld refinement results showed that increased Mg transformed the LaNi5phase in the La0.74Sm0.03Y0.23Ni4.32Al0.04precursor alloy into a superlattice *** electron microscopy（SEM） images showed that Mg was evenly distributed in the alloy *** in the superlattice significantly inhibited the phase decomposition of the superlattice structure during the hydrogen absorption/desorption *** A2B7-type La0.57Sm0.02Y0.18Mg0.23Ni3.38Al0.03alloy composed of Gd2Co7and Ce2Ni7phases was successfully *** pressure-composition isotherm profiles showed that the alloy had a hydrogen storage capacity as high as 1.73 wt%,with good cycling *** 50 cycles of hydrogen absorption/desorption,the alloy retained a hydrogen storage capacity of 1.45 wt%,with a capacity retention rate of up to 84.28%.The Mg thermal diffusion process thus provides a new approach for the controlled preparation of La-Mg-Ni-based alloys.

关键词： La–Mg–Ni-based hydrogen storage alloy A 2 B 7 -type superlattice alloy Mg thermal diffusion process Phase transformation Hydrogen storage property

Fabrication of Biomimetic Surface for Hydrophobic and Anti-icing Purposes via the Capillary Force Lithography

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Journal of Bionic Engineering 2024年第1期21卷 74-83页

作者： Wenqiang xing Yiping Tang Fengzhou Zhao Lichun zhang Dengying zhang School of Physics and Optoelectronic Engineering Ludong UniversityYantai 264025China

In this paper,inspired by lotus leaf surfaces,we fabricated biomimetic multi-scale micro-nano-structures by Two-Step Capillary Force Lithography(TS-CFL)and UV-assisted Capillary Force Lithography(UV-CFL).The experimental results indicated that TS-CFL was unfitted to fabricate large-area multi-scale ***,UV-CFL can fabricate large-area multi-scale *** discussed the hydrophobic and anti-icing properties of the biomimetic surfaces fabricated by these two *** found that small structures are significant for improving the hydrophobic anti-icing properties of single-structured or structureless *** believe that these results can complement the experimental details of both technologies and enable the development of more interesting micro-nano-structures biomimetic surfaces by both technologies in the future.

关键词： Capllary force lithography Biomimetic lotus leaf Multi-scale structures Hydrophobic Anti-icing

Determination of sulfite in water samples by flow injection analysis with fluorescence detection

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Chinese Chemical Letters 2010年第12期21卷 1457-1461页

作者： Li Qian Yin Dong xing Yuan min zhang State Key Laboratory of Marine Environmental Science Environmental Science Research Centre Xiamen University Xiamen 361005 China

A fast,sensitive,and reliable method for the determination of sulfite（SO3^2-） in fresh water and seawater samples was *** proposed method was based on the reaction of o-phthalaldehyde（OPA）-sulfite-NH_3 in alkaline solution,with flow injection analysis and fluorescence *** experimental parameters were investigated in pure water and seawater matrixes. The detection limits（S/N = 3） were 0.006μmol/L in pure water and 0.018μmol/L in seawater for SO3^2-.The method was successfully applied to analyze SO3^2- iin the samples of rain water and flue gas desulfurization seawater.

关键词： Sulfite Fluorophotometry FIA Flue gas desulfurization seawater Rainwater

维普期刊数据库

Multi-instance partial-label learning: towards exploiting dual inexact supervision

在线全文

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2024年第3期67卷 48-61页

作者： Wei TANG Weijia zhang min-Ling zhang School of Computer Science and Engineering Southeast University Key Laboratory of Computer Network and Information Integration (Southeast University) Ministry of Education School of Information and Physical Sciences The University of Newcastle

Weakly supervised machine learning algorithms are able to learn from ambiguous samples or labels, e.g., multi-instance learning or partial-label learning. However, in some real-world tasks, each training sample is associated with not only multiple instances but also a candidate label set that contains one groundtruth label and some false positive labels. Specifically, at least one instance pertains to the ground-truth label while no instance belongs to the false positive labels. In this paper, we formalize such problems as multi-instance partial-label learning(MIPL). Existing multi-instance learning algorithms and partial-label learning algorithms are suboptimal for solving MIPL problems since the former fails to disambiguate a candidate label set, and the latter cannot handle a multi-instance bag. To address these issues, a tailored algorithm named MIPLGP, i.e., multi-instance partial-label learning with Gaussian processes, is *** first assigns each instance with a candidate label set in an augmented label space, then transforms the candidate label set into a logarithmic space to yield the disambiguated and continuous labels via an exclusive disambiguation strategy, and last induces a model based on the Gaussian processes. Experimental results on various datasets validate that MIPLGP is superior to well-established multi-instance learning and partial-label learning algorithms for solving MIPL problems.

关键词： machine learning multi-instance partial-label learning multi-instance learning partial-label learning Gaussian processes

同方期刊数据库