检索结果-南通市图书馆

A Pipelining Loop Optimization Method for Dataflow Architecture

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 Computer Science & Technology 2018年第1期33卷 116-130页

作者： xu Tan Xiao-Chun Ye Xiao-Wei Shen yuan-chao xu Da Wang Lunkai Zhang Wen-Ming Li Dong-Rui Fan Zhi-Min Tang State Key Laboratory of Computer Architecture Institute of Computing Technology Chinese Academy of Sciences Beijing 100190 China School of Computer and Control Engineering University of Chinese Academy of Sciences Beijing 1000J9 China State Key Laboratory of Mathematical Engineering and Advanced Computing Wuxi 214125 China College of Information Engineering Capital Normal University Beijing 100048 China Department of Computer Science The University of Chicago Chicago IL 60637 U.S.A.

With the coming of exascale supercomputing era, power efficiency has become the most important obstacle to build an exascale system. Dataflow architecture has native advantage in achieving high power efficiency for scientific applications. However, the state-of-the-art dataflow architectures fail to exploit high parallelism for loop processing. To address this issue, we propose a pipelining loop optimization method （PLO）, which makes iterations in loops flow in the processing element （PE） array of dataflow accelerator. This method consists of two techniques, architecture-assisted hardware iteration and instruction-assisted software iteration. In hardware iteration execution model, an on-chip loop controller is designed to generate loop indexes, reducing the complexity of computing kernel and laying a good f（mndation for pipelining execution. In software iteration execution model, additional loop instructions are presented to solve the iteration dependency problem. Via these two techniques, the average number of instructions ready to execute per cycle is increased to keep floating-point unit busy. Simulation results show that our proposed method outperforms static and dynamic loop execution model in floating-point efficiency by 2.45x and 1.1x on average, respectively, while the hardware cost of these two techniques is acceptable.

关键词： dataflow model control-flow model loop optimization exascale computing scientific application

Numerical simulation of microstructure and microporosity morphology in directional solidification of aluminum-copper alloys:Effect of copper content and withdrawal rate

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

China Foundry 2025年第1期22卷 33-44页

作者： Wei yuan Hai-dong Zhao xu Shen Chun Zou yuan Liu Qing-yan xu National Engineering Research Center of Near-Net-Shape Forming for Metallic Materials South China University of TechnologyGuangzhou 510640China Key Laboratory for Advanced Materials Processing Technology(Ministry of Education) Tsinghua UniversityBeijing 100084China

Microporosity formed in the solidification process of Al alloys is detrimental to the alloy properties.A two-dimensional cellular automaton(CA)model was developed to simulate the microstructure and microporosity formation in Al-Cu alloys,considering variations in Cu content and solidification *** results indicate that the Cu content primarily influences the growth of *** validate the model,directional solidification experiments were conducted on Al-Cu alloys with varing Cu contents and withdrawal *** experimental results of dendrites and microporosity characteristics agree well with the predictions from the developed model,thus confirming the validity of the *** alloy’s liquidus temperature,dendrite morphology,and hydrogen saturation solubility arising from different Cu contents have significant effects on microporosity *** withdrawal rate primarily affects the nucleation of hydrogen microporosity by altering cooling rates and dendritic growth rates,resulting in different microporosity characteristics.

关键词： microporosity dendrites cellular automaton Al-Cu alloys directional solidification

维普期刊数据库

Support electron inductive effect of Pd-Mn/Ni foam catalyst for robust electrocatalytic hydrodechlorination

在线全文

学校读者我要写书评

暂无评论

Journal of Environmental Sciences 2025年第3期149卷 288-300页

作者： Junxi Li chao Feng Chong Chen yuan Pan Yunqi Liu State Key Laboratory of Heavy Oil Processing China University of Petroleum(East China)Qingdao 266580China Key Laboratory of Biofuels Qingdao Institute of Bioenergy and Bioprocess TechnologyChinese Academy of SciencesQingdao 266101China

Structural regulation of Pd-based electrocatalytic hydrodechlorination(EHDC)catalyst for constructing high-efficient cathode materials with low noble metal content and high atom utilization is crucial but still ***,a support electron inductive effect of Pd-Mn/Ni foam catalyst was proposed via in-situ Mn doping to optimize the electronic structure of the Ni foam(NF),which can inductive regulation of Pd for improving the EHDC *** mass activity and current efficiency of Pd-Mn/NF catalyst are 2.91 and 1.34 times superior to that of Pd/NF with 2,4-dichlorophenol as model compound,*** Mn-doped interlayer optimized the electronic structure of Pd by bringing the d-state closer to the Fermi level than Pd on the NF surface,which optimizied the binding of EHDC ***,the Mn-doped interlayer acted as a promoter for generating H∗and accelerating the EHDC *** work presents a simple and effective regulation strategy for constructing high-efficient cathode catalyst for the EHDC of chlorinated organic compounds.

关键词： Palladium Structural Regulation Electrocatalytic hydrodechlorination Support electron inductive effect

Tunable non-Markovian and quantum coherence in the single-qubit dephasing noise channel

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Science China(Physics,Mechanics & Astronomy) 2025年第3期68卷 92-101页

作者： Na-Na Zhang chao-Yi Wu xu Zhou Qi-yuan Liu Cheng-Ge Liu Yong-Rui Guo Ren-Pu Li School of Optoelectronic Engineering Chongqing University of Posts and TelecommunicationsChongqing400065China Department of Physics Applied Optics Beijing Area Major LaboratoryBeijing Normal UniversityBeijing100875China

In this paper,we construct a single-qubit dephasing noise channel based on the nuclear magnetic resonance(NMR)system by employing the bath-engineering technology,and achieve the construction of the tunable non-Markovian environment in the dephasing noise *** findings indicate that for the single-qubit system,the transition of system dynamics from Markovian to non-Markovian can be achieved by adjusting the base frequency of the noise power ***,the base frequency corresponding to this phase transition point is not fixed,and there is a certain relationship between it and the total evolution time of the single-qubit *** our research,we discovered a fundamental relationship:if the single-qubit system dynamics undergoe a transition from Markovian to non-Markovian atω_(0) within 0-2t ms,shortening the evolution time to 0-t ms results in an increase of the phase transition point to 2ω_(0).This insight offers crucial guidance for artificially crafting non-Markovian environments across arbitrary time scales in single-qubit systems,and it is not limited by the type of *** from system dynamics,quantum coherence is also a focal point of our *** find that when the system dynamics exhibit non-Markovian behavior,the quantum coherence of the single-qubit system experiences ***,the timing of these coherence revivals aligns with the instants of the non-Markovianity ***,our research also serves as a pivotal foundation for the artificial manipulation and realization of quantum coherence revivals within diverse single-qubit systems.

关键词： dephasing noise channel non-Markovian quantum coherence

维普期刊数据库

同方期刊数据库

Decentralized Federated Learning Algorithm Under Adversary Eavesdropping

在线全文

学校读者我要写书评

暂无评论

IEEE/CAA Journal of Automatica Sinica 2025年第2期12卷 448-456页

作者： Lei xu Danya xu Xinlei Yi chao Deng Tianyou Chai Tao Yang the State Key Laboratory of Synthetical Automation for Process Industries Northeastern University the Department of Mechanical Engineering University of Victoria Shanghai Institute of Intelligent Science and Technology Tongji University IEEE the Institute of Advanced Technology Nanjing University of Posts and Telecommunications

In this paper, we study the decentralized federated learning problem, which involves the collaborative training of a global model among multiple devices while ensuring data *** classical federated learning, the communication channel between the devices poses a potential risk of compromising private information. To reduce the risk of adversary eavesdropping in the communication channel, we propose TRADE(transmit difference weight) concept. This concept replaces the decentralized federated learning algorithm's transmitted weight parameters with differential weight parameters, enhancing the privacy data against eavesdropping. Subsequently, by integrating the TRADE concept with the primal-dual stochastic gradient descent(SGD)algorithm, we propose a decentralized TRADE primal-dual SGD algorithm. We demonstrate that our proposed algorithm's convergence properties are the same as those of the primal-dual SGD algorithm while providing enhanced privacy protection. We validate the algorithm's performance on fault diagnosis task using the Case Western Reserve University dataset, and image classification tasks using the CIFAR-10 and CIFAR-100 datasets,revealing model accuracy comparable to centralized federated learning. Additionally, the experiments confirm the algorithm's privacy protection capability.

关键词： Adversary eavesdropping decentralized federated learning privacy protection stochastic gradient descent

Oxygen vacancies enriched Ir/WOx catalysts for the directly chem-catalytic conversion of cellulose to ethanol

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Tungsten 2025年第1期7卷 88-99页

作者： Yu-Jing Weng Zhao-Ying Ding Ying-chao Li yuan-Feng Wu yuan-Yang xu Rui Chen Xiao-Lei Zhao Hai-Yong Wang Da-Lei Zhang Yu-Long Zhang Henan Key Laboratory of Coal Green Conversion and International Joint Laboratory of Coal Clean Utilization College of Chemistry and Chemical Engineering Henan Polytechnic University Key Laboratory of Renewable Energy Guangzhou Institute of Energy Conversion Chinese Academy of Sciences Liaoning Institute of Energy Technology

Tungsten and its oxides are highly versatile catalysts with unique properties that hold great promise for a wide array of applications,including biological pathways and chemical *** chemo-catalysis of cellulose to ethanol presents a promising avenue for ethanol synthesis with the potential for enhanced efficiency compared to conventional methods,offering a sustainable and carbon-neutral alternative to existing petrochemical and biological ***,Ir/WOxcatalysts enriched with oxygen vacancies were meticulously prepared and demonstrated remarkable catalytic performance in converting cellulose to ethanol directly in the presence of tungstic *** catalyst achieved a high ethanol yield of 66.5%under optimal reaction conditions,approaching the theoretical yield limit of traditional ethanol *** results revealed the presence of oxygen defects,Bronsted/Lewis acidity,and Ir nanoclusters on the catalyst surface,working synergistically to facilitate the cascade catalysis of cellulose hydrolysis,sugar retro-aldol condensation reaction,and C2 intermediate hydrodeoxygenation to *** situ Fourier transform infrared spectrometer analysis confirmed the catalyst's ability to enhance the adsorption and activation of ethylene glycol for ethanol *** manipulating subtle structural features,this study offers a fresh perspective on tungsten group catalyst design and underscores the importance of synergistic effects among catalytic sites for the efficient production of ethanol from cellulose.

关键词： Cellulosic ethanol Chem-catalysis Hydrogenolysis Retro-aldol condensation Tungsten

Graphene/F_(16)CuPc synaptic transistor for the emulation of multiplexed neurotransmission

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 Semiconductors 2025年第1期46卷 215-223页

作者： Zhipeng xu Yao Ni Mingxin Sun Yiming yuan Ning Wu Wentao xu Institute of Photoelectronic Thin Film Devices and Technology Key Laboratory of Photoelectronic Thin Film Devices and Technology of TianjinCollege of Electronic Information and Optical EngineeringEngineering Research Center of Thin Film Photoelectronic Technology of Ministry of EducationNankai UniversityTianjin 300350China Shenzhen Research Institute of Nankai University Shenzhen 518000China Smart Sensing Interdisciplinary Science Center Nankai UniversityTianjin 300350China School of Integrated Circuits Guangdong University of TechnologyGuangzhou 510006China

We demonstrate a bipolar graphene/F_(16)CuPc synaptic transistor(GFST)with matched p-type and n-type bipolar properties,which emulates multiplexed neurotransmission of the release of two excitatory neurotransmitters in graphene and F_(16)CuPc channels,*** process facilitates fast-switching plasticity by altering charge carriers in the separated *** complementary neural network for image recognition of Fashion-MNIST dataset was constructed using the matched relative amplitude and plasticity properties of the GFST dominated by holes or electrons to improve the weight regulation and recognition accuracy,achieving a pattern recognition accuracy of 83.23%.These results provide new insights to the construction of future neuromorphic systems.

关键词： synaptic transistor dual excitatory channels fast-switching plasticity multiplexed neurotransmission

维普期刊数据库

Direct Synthesis of UOS Zeolite Using a Simple and Commercially Available Organic Structure-directing Agent

在线全文

学校读者我要写书评

暂无评论

Chemical Research in Chinese Universities 2025年第1期41卷 113-120页

作者： HU chao FU Wenhua yuan Zhiqing LIU Chuang WANG Zhendong YANG Weimin School of Chemical Engineering East China University of Science and TechnologyShanghai200237P.R.China State Key Laboratory of Green Chemical Engineering and Industrial Catalysis Sinopec Shanghai Research Institute of Petrochemical Technology Co.Ltd.SinopecShanghai201208P.R.China

The preparation of zeolite utilizing commercially available organic compounds instead of complex and expensive ones is of practical ***,we report the synthesis of germanosilicate zeolite with UOS framework by utilizing a simple and commercially available compound 3-diethylamino-1-propanol(DEAP)as organic structure-directing agent(OSDA)under fluoride *** synthesis has been optimized by rational modification of the variables,including the Si/Ge molar ratios,the amount of DEAP and F−anions,the concentration of the synthesis gel and crystallization *** zeolite materials were prepared with Si/Ge ratio in the range of 1–*** physicochemical properties,including crystallinity,crystal morphology,chemical environment of framework elements,textural properties and acidity were characterized by multiple *** atoms are proved to preferentially occupy the T sites in the double-four-ring(D4Rs)*** to the isostructural IM-16 zeolite,the UOS zeolites prepared herein are of similar textural properties,such as specific surface area and micropore *** simple structure and commercial availability of DEAP endow this synthesis with a cost advantage over the conventional preparation of UOS zeolite,where an expensive imidazolium derivative is employed.

关键词： UOS zeolite Organic structure-directing agent Hydrothermal synthesis Physicochemical property

A Deep Learning Estimation Method for Temperature-Induced Girder End Displacements of Suspension Bridges

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Structural Durability & Health Monitoring 2025年第2期19卷 307-325页

作者： Yao Jin yuan Ren Chong-yuan Guo Chong Li Zhao-yuan Guo Xiang xu School of Transportation Southeast UniversityNanjing210096China Department of Engineering Guangzhou Expressway Operation Management Co.Ltd.Guangzhou510000China Department of Engineering CCCC Highway Bridge National Engineering Research Center Co.Ltd.Beijing100088China Department of Engineering Jiangsu Provincial Transportation Engineering Construction BureauNanjing210004China

To improve the accuracy of thermal response estimation and overcome the limitations of the linear regression model and Artificial Neural Network(ANN)model,this study introduces a deep learning estimation method specifically based on the Long Short-Term Memory(LSTM)network,to predict temperature-induced girder end displacements of the Dasha Waterway Bridge,a suspension bridge in ***,to enhance data quality and select target sensors,preprocessing based on the sigma rule and nearest neighbor interpolation is applied to the raw ***,to eliminate the high-frequency components from the displacement signal,the wavelet transform is ***,the linear regression model and ANN model are established,whose results do not meet the requirements and fail to address the time lag effect between temperature and *** study proceeds to develop the LSTM network model and determine the optimal parameters through hyperparameter sensitivity ***,the results of the LSTM network model are discussed by a comparative analysis against the linear regression model and ANN model,which indicates a higher accuracy in predicting temperatureinduced girder end displacements and the ability to mitigate the time-lag *** be more specific,in comparison between the linear regression model and LSTM network,the mean square error decreases from 6.5937 to 1.6808 and R2 increases from 0.683 to 0.930,which corresponds to a 74.51%decrease in MSE and a 36.14%improvement in *** to ANN,with an MSE of 4.6371 and an R2 of 0.807,LSTM shows a decrease in MSE of 63.75%and an increase in R2 of 13.23%,demonstrating a significant enhancement in predictive performance.

关键词： Suspension bridges thermal response girder end displacement deep learning