检索结果-南通市图书馆

Solid Desiccant Heat Pump Fresh Air Unit using Composite Silica Gel

在线全文

学校读者我要写书评

暂无评论

Journal of Thermal Science 2024年第4期33卷 1286-1300页

作者： yang Tianyu GUO Defang GE Tianshu Institute of Refrigeration and Cryogenics Shanghai Jiao Tong UniversityKey Laboratory of Power Mechanical EngineeringMOE ChinaShanghai200240China Qingdao Haier A/C Electronic Co. LtdQingdao266101China

The introduction of fresh air into the indoor space leads to a significant increase in cooling or heating *** desiccant heat pump fresh air unit which can handle the latent and sensible load of fresh air efficiently have been proposed *** improve the performance of the solid desiccant heat pump fresh air unit in the fresh air handling process,in this paper,the application of composite silica gel in a heat pump fresh air unit was *** comparison between silica gel coating(SGC)and composite silica gel coating(CSGC)shows that the adsorption rate and water uptake capacity of CSGC are more than two times higher than those of *** experimental setup for the solid desiccant heat pump fresh air unit was *** performance of SGC and CSGC was tested in the setup *** show that under summer conditions,compared with the solid desiccant heat pump fresh air unit using silica gel(SGFU),the average moisture removal and COP of the one using composite silica gel(CSGFU)increased by 15%and 30%,*** winter conditions,compared with SGFU,the average humidification and COP of CSGFU increased by 42%and 17%.With optimal operation conditions of 3 min switchover time and 40 r/s compressor frequency,the COP of CSGFU under summer conditions can reach *** also show that the CSGFU and SGFU have higher COP and dehumidification rate under higher outdoor temperature and humidity ratio.

关键词： fresh air system desiccant coated heat exchanger humidification dehumidification experimental comparison

Network Configuration Entity Extraction Method Based on Transformer with Multi-Head Attention Mechanism

在线全文

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2024年第1期78卷 735-757页

作者： yang yang Zhenying Qu Zefan Yan Zhipeng Gao Ti Wang State Key Laboratory of Networking and Switching Technology Beijing University of Posts and TelecommunicationsBeijing100876China Product Development Department China Unicom Smart City Research InstituteBeijing100044China

Nowadays,ensuring thequality of networkserviceshas become increasingly *** are turning toknowledge graph technology,with a significant emphasis on entity extraction in the identification of device *** research paper presents a novel entity extraction method that leverages a combination of active learning and attention ***,an improved active learning approach is employed to select the most valuable unlabeled samples,which are subsequently submitted for expert *** approach successfully addresses the problems of isolated points and sample redundancy within the network configuration sample *** the labeled samples are utilized to train the model for network configuration entity ***,the multi-head self-attention of the transformer model is enhanced by introducing the Adaptive Weighting method based on the Laplace mixture *** enhancement enables the transformer model to dynamically adapt its focus to words in various positions,displaying exceptional adaptability to abnormal data and further elevating the accuracy of the proposed *** comparisons with Random Sampling(RANDOM),Maximum Normalized Log-Probability(MNLP),Least Confidence(LC),Token Entrop(TE),and Entropy Query by Bagging(EQB),the proposed method,Entropy Query by Bagging and Maximum Influence Active Learning(EQBMIAL),achieves comparable performance with only 40% of the samples on both datasets,while other algorithms require 50% of the ***,the entity extraction algorithm with the Adaptive Weighted Multi-head Attention mechanism(AW-MHA)is compared with BILSTM-CRF,Mutil_Attention-Bilstm-Crf,Deep_Neural_Model_NER and BERT_Transformer,achieving precision rates of 75.98% and 98.32% on the two datasets,*** tests demonstrate the statistical significance and effectiveness of the proposed algorithms in this paper.

关键词： Entity extraction network configuration knowledge graph active learning transformer

Untethered Micro/Nanorobots for Remote Sensing:Toward Intelligent Platform

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-Micro Letters 2024年第2期16卷 450-483页

作者： Qianqian Wang Shihao yang Li Zhang Jiangsu Key Laboratory for Design and Manufacture of Micro‑Nano Biomedical Instruments School of Mechanical EngineeringSoutheast UniversityNanjing 211189People’s Republic of China Department of Mechanical and Automation Engineering The Chinese University of Hong KongShatinHong Kong 999077People’s Republic of China Chow Yuk Ho Technology Centre for Innovative Medicine The Chinese University of Hong KongShatinHong Kong 999077People’s Republic of China T Stone Robotics Institute The Chinese University of Hong KongShatinHong Kong 999077People’s Republic of China Department of Surgery The Chinese University of Hong KongShatinHong Kong 999077People’s Republic of China

Untethered micro/nanorobots that can wirelessly control their motion and deformation state have gained enormous interest in remote sensing applications due to their unique motion characteristics in various media and diverse *** are developing micro/nanorobots as innovative tools to improve sensing performance and miniaturize sensing systems,enabling in situ detection of substances that traditional sensing methods struggle to *** the past decade of development,significant research progress has been made in designing sensing strategies based on micro/nanorobots,employing various coordinated control and sensing *** review summarizes the latest developments on micro/nanorobots for remote sensing applications by utilizing the self-generated signals of the robots,robot behavior,microrobotic manipulation,and robot-environment *** recent studies and relevant applications in remote sensing,we also discuss the challenges and future perspectives facing micro/nanorobots-based intelligent sensing platforms to achieve sensing in complex environments,translating lab research achievements into widespread real applications.

关键词： Micro/nanorobot Remote sensing Wireless control Self-propulsion Actuation at small scales

Intelligent feedforward gust alleviation based on neural network

在线全文

学校读者我要写书评

暂无评论

Chinese Journal of Aeronautics 2024年第3期37卷 116-132页

作者： Yitao ZHOU Zhigang WU Chao yang School of Aeronautic Science and Engineering Beihang UniversityBeijing 100191China

This paper proposes a neural network-based intelligent feedforward gust alleviation framework,which includes a neural network identification model and a neural network controller.A neural network training dataset is formed by collecting flight data and the gust data encountered during the aircraft flight.A neural network identification model is first trained to accurately predict the aircraft’s ***,based on the output of the identification model and the collected flight data,the parameters of the time-delay neural network controller are obtained through a learning *** simulation results show that the designed intelligent controller has good gust alleviation effects for both continuous turbulence excitation and discrete gust *** example,when the aircraft is 40000 kg and the flight speed is 0.81Ma,the controller achieves a 67.82%reduction in wingtip acceleration and a 35.90%reduction in center of mass acceleration under continuous turbulence *** considering the measurement uncertainties,such as noise existing in the collected data,the trained controller can still achieve an acceptable gust alleviation ***,considering a flight in which the aircraft mass is constantly changing,the intelligent controller,which continuously learns from new flight data,maintains a good gust alleviation effect throughout the flight.

关键词： Gust alleviation Intelligent control Feedforward control Neural networks Time-varying aircraft

Understanding the Interfacial Energy Structure and Electron Extraction Process in Inverted Organic Solar Cells with Phosphine-Doped Cathode Interlayers

在线全文

学校读者我要写书评

暂无评论

Chinese Journal of Chemistry 2024年第12期42卷 1347-1354页

作者： Yi yang Jingwen Wang yang Xiao Bowei Xu Jianhui Hou State Key Laboratory of Polymer Physics and Chemistry Beijing National Laboratory for Molecular SciencesInstitute of ChemistryChinese Academy of SciencesBeijing 100190China State Key Laboratory of Chemical Resource Engineering College of Materials Science and EngineeringBeijing University of Chemical TechnologyBejing 100029China University of Chinese Academy of Sciences Beijing 100049China

Comprehensive Summary Cathode interlayers(CILs)play an essential role in achieving efficient organic solar cells(OSCs).However,the electronic structure at the electrode/CIL/active layer interfaces and the underlying mechanisms for electron collection remain unclear,which becomes a major obstacle to develop high-performance ***,we investigate the relationship of the electron collection abilities of four cross-linked and n-doped CILs(c-NDI:P0,c-NDI:P1,c-NDI:P2,c-NDI:P3)with their electronic structure of space charge region at heterojunction *** accurately calculating the depletion region width according to the barrier height,doping density and permittivity,we put forward that the optimal thickness of CIL should be consistent with the depletion region width to realize the minimum energy *** a result,the depletion region width is largely reduced from 13 nm to 0.8 nm at the indium tin oxide(ITO)/c-NDI:P0 interface,resulting in a decent PCE of 17.7%for the corresponding inverted OSCs.

关键词： Organic solar cells Cathode interlayer Depletion region width Electric conductivity Electron transport Electron transfer Electron extraction process

Active flutter suppression of nonlinear fin-actuator system via inverse system and immersion and invariance method

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 Journal of Aeronautics 2024年第4期37卷 343-362页

作者： Jin LU Zhigang WU Chao yang Chinese Aeronautical Establishment Beijing 100029China School of Aeronautic Science and Engineering Beihang UniversityBeijing 100191China

Structural nonlinearities such as freeplay will affect the stability and even flight safety of the fin-actuator *** is a lack of a practical method for designing Active Flutter Suppression (AFS) control laws for nonlinear fin-actuator systems.A design method for the AFS controller of the nonlinear all-movable fin-electromechanical actuator system is established by combining the inverse system and the Immersion and Invariance (I&I) ***,the composite control law combining the inverse system principle and internal model control is used to offset the nonlinearity and dynamics of the actuator,so that its driving torque can follow the ideal ***,the ideal torque of the actuator is designed employing the I&I *** unfavorable oscillation of the fin is suppressed by making the output torque of the actuator track the ideal *** simulation results reveal that the proposed AFS method can increase the flutter speed of the nonlinear finactuator system with freeplay,and a set of controller parameters is also applicable for wider freeplay within a certain *** power required for the actuator does not exceed the power that can be provided by the commonly used aviation *** method can also resist a certain level of noise and external disturbance.

关键词： Aeroelasticity Active flutter suppression Fin-actuator system Freeplay Inverse system Immersion and invariance

Reinforcement Learning Based Quantization Strategy Optimal Assignment Algorithm for Mixed Precision

在线全文

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2024年第4期79卷 819-836页

作者： Yuejiao Wang Zhong Ma Chaojie yang Yu yang Lu Wei Xi’an Microelectronics Technology Institute Xi’an710065China

The quantization algorithm compresses the original network by reducing the numerical bit width of the model,which improves the computation speed. Because different layers have different redundancy and sensitivity to databit width. Reducing the data bit width will result in a loss of accuracy. Therefore, it is difficult to determinethe optimal bit width for different parts of the network with guaranteed accuracy. Mixed precision quantizationcan effectively reduce the amount of computation while keeping the model accuracy basically unchanged. In thispaper, a hardware-aware mixed precision quantization strategy optimal assignment algorithm adapted to low bitwidth is proposed, and reinforcement learning is used to automatically predict the mixed precision that meets theconstraints of hardware resources. In the state-space design, the standard deviation of weights is used to measurethe distribution difference of data, the execution speed feedback of simulated neural network accelerator inferenceis used as the environment to limit the action space of the agent, and the accuracy of the quantization model afterretraining is used as the reward function to guide the agent to carry out deep reinforcement learning training. Theexperimental results show that the proposed method obtains a suitable model layer-by-layer quantization strategyunder the condition that the computational resources are satisfied, and themodel accuracy is effectively *** proposed method has strong intelligence and certain universality and has strong application potential in thefield of mixed precision quantization and embedded neural network model deployment.

关键词： Mixed precision quantization quantization strategy optimal assignment reinforcement learning neural network model deployment

Stochastic Computing Convolutional Neural Network Architecture Reinvented for Highly Efficient Artificial Intelligence Workload on Field-Programmable Gate Array

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Research 2024年第2期2024卷 55-79页

作者： yang yang Lee Zaini Abdul Halim Mohd Nadhir Ab Wahab Tarik Adnan Almohamad School of Electrical and Electronic Engineering Universiti Sains MalaysiaNibong Tebal 14300 Penang Malaysia School of Computer Sciences Universiti Sains MalaysiaGelugor11800 PenangMalaysia Electrical-Electronics Engineering Department Faculty of EngineeringKarabuk University78050 KarabukTurkiye

Stochastic computing(SC)has a substantial amount of study on application-specific integrated circuit(ASIC)design for artificial intelligence(AI)edge computing,especially the convolutional neural network(CNN)***,SC has little to no optimization on field-programmable gate array(FPGA).Scaling up the ASIC logic without FPGA-oriented designs is inefficient,while aggregating thousands of bitstreams is still challenging in the conventional *** research has reinvented several FPGA-efficient 8-bit SC CNN computing architectures,i.e.,SC multiplexer multiply-accumulate,multiply-accumulate function generator,and binary rectified linear unit,and successfully scaled and implemented a fully parallel CNN model on Kintex7 *** proposed SC hardware only compromises 0.14%accuracy compared to binary computing on the handwriting Modified National Institute of Standards and Technology classification task and achieved at least 99.72%energy saving per image feedforward and 31?more data throughput than modern *** to SC,early decision termination pushed the performance baseline exponentially with minimum accuracy loss,making SC CNN extremely lucrative for AI edge computing but limited to classification *** SC's inherent noise heavily penalizes CNN regression performance,rendering SC unsuitable for regression tasks.

关键词： hardware Highly rendering

The Lightweight Edge-Side Fault Diagnosis Approach Based on Spiking Neural Network

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6期79卷 4883-4904页

作者： Jingting Mei yang yang Zhipeng Gao Lanlan Rui Yijing Lin State Key Laboratory of Networking and Switching Technology Beijing University of Posts and TelecommunicationsBeijing100876China

Network fault diagnosis methods play a vital role in maintaining network service quality and enhancing user experience as an integral component of intelligent network *** the unique characteristics of edge networks,such as limited resources,complex network faults,and the need for high real-time performance,enhancing and optimizing existing network fault diagnosis methods is ***,this paper proposes the lightweight edge-side fault diagnosis approach based on a spiking neural network(LSNN).Firstly,we use the Izhikevich neurons model to replace the Leaky Integrate and Fire(LIF)neurons model in the LSNN *** neurons inherit the simplicity of LIF neurons but also possess richer behavioral characteristics and flexibility to handle diverse data *** by Fast Spiking Interneurons(FSIs)with a high-frequency firing pattern,we use the parameters of ***,inspired by the connection mode based on spiking dynamics in the basal ganglia(BG)area of the brain,we propose the pruning approach based on the FSIs of the BG in LSNN to improve computational efficiency and reduce the demand for computing resources and energy ***,we propose a multiple iterative Dynamic Spike Timing Dependent Plasticity(DSTDP)algorithm to enhance the accuracy of the LSNN *** on two server fault datasets demonstrate significant precision,recall,and F1 improvements across three diagnosis ***,lightweight indicators such as Params and FLOPs significantly reduced,showcasing the LSNN’s advanced performance and model *** conclude,experiment results on a pair of datasets indicate that the LSNN model surpasses traditional models and achieves cutting-edge outcomes in network fault diagnosis tasks.

关键词： Network fault diagnosis edge networks Izhikevich neurons pruning dynamic spike timing dependent plasticity learning