检索结果-南通市图书馆

A probabilistic generative model for tracking multi-knowledge concept mastery probability

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Frontiers of Computer Science 2024年第3期18卷 155-170页

作者： Hengyu LIU Tiancheng zhang fan LI Minghe YU Ge YU School of Computer Science and Engineering Northeastern UniversityShenyang 110169China Software College Northeastern UniversityShenyang 110169China

Knowledge tracing aims to track students’knowledge status over time to predict students’future performance *** a real environment,teachers expect knowledge tracing models to provide the interpretable result of knowledge *** chain-based knowledge tracing(MCKT)models,such as Bayesian Knowledge Tracing,can track knowledge concept mastery probability over ***,as the number of tracked knowledge concepts increases,the time complexity of MCKT predicting student performance increases exponentially(also called explaining away problem).When the number of tracked knowledge concepts is large,we cannot utilize MCKT to track knowledge concept mastery probability over *** addition,the existing MCKT models only consider the relationship between students’knowledge status and problems when modeling students’responses but ignore the relationship between knowledge concepts in the same *** address these challenges,we propose an inTerpretable pRobAbilistiC gEnerative moDel(TRACED),which can track students’numerous knowledge concepts mastery probabilities over *** solve explain away problem,we design long and short-term memory(LSTM)-based networks to approximate the posterior distribution,predict students’future performance,and propose a heuristic algorithm to train LSTMs and probabilistic graphical model *** better model students’exercise responses,we proposed a logarithmic linear model with three interactive strategies,which models students’exercise responses by considering the relationship among students’knowledge status,knowledge concept,and *** conduct experiments with four real-world datasets in three knowledge-driven *** experimental results show that TRACED outperforms existing knowledge tracing methods in predicting students’future performance and can learn the relationship among students,knowledge concepts,and problems from students’exercise *** also conduct several case *** case studies show that

关键词： probabilistic graphical model deep learning knowledge tracing learner modeling

Vibration Control of the Rail Grinding Vehicle with Abrasive Belt Based on Structural Optimization and Lightweight Design

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 Mechanical Engineering 2024年第3期37卷 311-337页

作者： Wengang fan Shuai zhang zhiwei Wu Yi Liu Jiangnan Yu School of Mechanical Electronic and Control EngineeringBeijing Jiaotong UniversityBeijing 100044China Key Laboratory of Vehicle Advanced Manufacturing Measuring and Control TechnologyMinistry of EducationBeijing 100044China China Railway Materials Operation and Maintenance Technology Co. LtdBeijing 100036China

As a new grinding and maintenance technology,rail belt grinding shows significant advantages in many applications The dynamic characteristics of the rail belt grinding vehicle largely determines its grinding performance and service *** order to explore the vibration control method of the rail grinding vehicle with abrasive belt,the vibration response changes in structural optimization and lightweight design are respectively analyzed through transient response and random vibration simulations in this ***,the transient response simulation analysis of the rail grinding vehicle with abrasive belt is carried out under operating conditions and non-operating ***,the vibration control of the grinding vehicle is implemented by setting vibration isolation elements,optimizing the structure,and increasing ***,in order to further explore the dynamic characteristics of the rail grinding vehicle,the random vibration simulation analysis of the grinding vehicle is carried out under the condition of the horizontal irregularity of the American AAR6 ***,by replacing the Q235 steel frame material with 7075 aluminum alloy and LA43M magnesium alloy,both vibration control and lightweight design can be achieved *** results of transient dynamic response analysis show that the acceleration of most positions in the two working conditions exceeds the standard value in GB/T 17426-1998 *** optimizing the structure of the grinding vehicle in three ways,the average vibration acceleration of the whole car is reduced by about 55.1%from 15.6 m/s^(2) to 7.0 m/s^(2).The results of random vibration analysis show that the grinding vehicle with Q235 steel frame does not meet the safety conditions of 3σ.By changing frame material,the maximum vibration stress of the vehicle can be reduced from 240.7 MPa to 160.0 MPa and the weight of the grinding vehicle is reduced by about 21.7%from 1500 kg to 1175 *** modal analysis results indic

关键词： Vibration control Dynamic characteristics Structural optimization Lightweight design Modal analysis

Data complexity-based batch sanitization method against poison in distributed learning

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Digital Communications and Networks 2024年第2期10卷 416-428页

作者： Silv Wang Kai fan Kuan zhang Hui Li Yintang Yang State Key Laboratory of Integrated Service Networks Xidian UniversityXi'an710126China Department of Electrical and Computer Engineering University of Nebraska-LincolnLincolnNE68588USA Key Lab.of the Minist.of Educ.for Wide Bandgap Semiconductor Materials and Devices Xidian UniversityXi'an710071China

The security of Federated Learning(FL)/Distributed Machine Learning(DML)is gravely threatened by data poisoning attacks,which destroy the usability of the model by contaminating training samples,so such attacks are called causative availability indiscriminate *** the problem that existing data sanitization methods are hard to apply to real-time applications due to their tedious process and heavy computations,we propose a new supervised batch detection method for poison,which can fleetly sanitize the training dataset before the local model *** design a training dataset generation method that helps to enhance accuracy and uses data complexity features to train a detection model,which will be used in an efficient batch hierarchical detection *** model stockpiles knowledge about poison,which can be expanded by retraining to adapt to new *** neither attack-specific nor scenario-specific,our method is applicable to FL/DML or other online or offline scenarios.

关键词： Distributed machine learning security Federated learning Data poisoning attacks Data sanitization Batch detection Data complexity

Distributed power conditioning unit of large-scale space solar power station

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8期37卷 421-434页

作者： Xintong LI Jianwei MI Yiqun zhang Guanheng fan Jie DU School of Mechano-electronic Engineering Xidian UniversityXi’an 710071China Shaanxi Key Laboratory of Space Solar Power Station System Xidian UniversityXi’an 710071China Academy of Advanced Interdisciplinary Research Xidian UniversityXi’an 710071China

In this paper,a multi-bus distributed Power Conditioning Unit(PCU)is proposed for the Space Solar Power Station with large scale photovoltaic(PV)array and power levels reaching MW *** this unit,there are multiple independent PV *** each PV array,there are multiple independent PV *** this paper,a V-P droop control method with adaptive droop coefficient is proposed,which modifies the droop intercept based on the bus voltage deviation and the power per unit value of the PV *** method ensures the accuracy of bus voltage and achieves proportional distribution of power between PV arrays based on the proposed topology structure in this *** the load changes or the output power of the PV array fluctuates,this method can ensure that power is distributed *** principle and control method of the proposed droop control method is analyzed in this *** effectiveness of the method is verified through MATLAB/Simulink simulation and *** and experimental results show that the proposed method can achieve power distributed proportionally when load changes and PV output power fluctuates,reduce bus voltage error caused by line impedance and differences in rated power of different PV arrays,and improve the performance of PV power generation system applied to space.

关键词： Space Solar Power Station Photovoltaics Droop control DC confluence Power conditioning unit

Cooperative decision-making algorithm with efficient convergence for UCAV formation in beyond-visual-range air combat based on multi-agent reinforcement 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

Chinese Journal of Aeronautics 2024年第8期37卷 311-328页

作者： Yaoming ZHOU fan YANG Chaoyue zhang Shida LI Yongchao WANG School of Aeronautic Science and Engineering Beihang UniversityBeijing 100191China Key Laboratory of Industrial Control Technology Institute of Cyber-Systems and ControlZhejiang UniversityHangzhou 310027China

Highly intelligent Unmanned Combat Aerial Vehicle(UCAV)formation is expected to bring out strengths in Beyond-Visual-Range(BVR)air *** Multi-Agent Reinforcement Learning(MARL)shows outstanding performance in cooperative decision-making,it is challenging for existing MARL algorithms to quickly converge to an optimal strategy for UCAV formation in BVR air combat where confrontation is complicated and reward is extremely sparse and *** to solve this problem,this paper proposes an Advantage Highlight Multi-Agent Proximal Policy Optimization(AHMAPPO)***,at every step,the AHMAPPO records the degree to which the best formation exceeds the average of formations in parallel environments and carries out additional advantage sampling according to ***,the sampling result is introduced into the updating process of the actor network to improve its optimization ***,the simulation results reveal that compared with some state-of-the-art MARL algorithms,the AHMAPPO can obtain a more excellent strategy utilizing fewer sample episodes in the UCAV formation BVR air combat simulation environment built in this paper,which can reflect the critical features of BVR air *** AHMAPPO can significantly increase the convergence efficiency of the strategy for UCAV formation in BVR air combat,with a maximum increase of 81.5%relative to other algorithms.

关键词： Unmanned combat aerial vehicle(UCAV)formation Decision-making Beyond-visual-range(BVR)air combat Advantage highlight Multi-agent reinforcement learning(MARL)

Solvent-free carbon sphere nanofluids towards intelligent lubrication regulation

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Friction 2024年第1期12卷 95-109页

作者： Sheng zhang Wen LI Xiaoliang MA Xiaoqiang fan Minhao ZHU Tribology Research Institute School of Mechanical EngineeringSouthwest Jiaotong UniversityChengdu 610031China Key Laboratory of Advanced Technologies of Materials Ministry of EducationSchool of Materials Science and EngineeringSouthwest Jiaotong UniversityChengdu 610031China

By simply switching the electrical circuit installed on steel/steel contact,the tribological behaviors of nanofluids(NFs)can be regulated in real time,thereby achieving the desired performance of friction reduction and wear ***,solvent-free carbon spherical nanofluids(C-NFs)were successfully prepared for intelligent lubrication regulation.C-NFs with excellent lubrication performance can immediately reduce the coefficient of friction(COF)despite applying a weak electric potential(1.5 V).Moreover,polyethylene glycol 400(PEG400)containing 5.0 wt%C-NFs remained responsive to electrical stimulation under the intermittent voltage application with an average coefficient of friction(ACOF)reduction of 20.8%over *** intelligent lubrication regulation of C-NFs under an external electric field(EEF)mainly depends on the orderly arranged double-electric adsorption film of ion canopy-adsorbed carbon spheres(CSs).The intermittent electrical application can continuously reinforce the adsorption film and its durability for real-time controlling the sliding ***-stimulation-responsive intelligent lubricants provide a new technical support for realizing intelligent stepless control of devices.

关键词： solvent-free carbon spherical nanofluids(C-NFs) intelligent lubrication regulation lubrication performance adsorption film

Unsupervised Multi-Expert Learning Model for Underwater Image Enhancement

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

IEEE/CAA Journal of Automatica Sinica 2024年第3期11卷 708-722页

作者： Hongmin Liu Qi zhang Yufan Hu Hui Zeng Bin fan the School of Intelligence Science and Technology and the Institute of Artificial Intelligence University of Science and Technology BeijingBeijing 100083China the School of Automation and Electrical Engineering University of Science and Technology BeijingBeijing 100083China IEEE

Underwater image enhancement aims to restore a clean appearance and thus improves the quality of underwater degraded *** methods feed the whole image directly into the model for ***,they ignored that the R,G and B channels of underwater degraded images present varied degrees of degradation,due to the selective absorption for the *** address this issue,we propose an unsupervised multi-expert learning model by considering the enhancement of each color ***,an unsupervised architecture based on generative adversarial network is employed to alleviate the need for paired underwater *** on this,we design a generator,including a multi-expert encoder,a feature fusion module and a feature fusion-guided decoder,to generate the clear underwater ***,a multi-expert discriminator is proposed to verify the authenticity of the R,G and B channels,*** addition,content perceptual loss and edge loss are introduced into the loss function to further improve the content and details of the enhanced *** experiments on public datasets demonstrate that our method achieves more pleasing results in vision *** metrics(PSNR,SSIM,UIQM and UCIQE) evaluated on our enhanced images have been improved obviously.

关键词： Multi-expert learning underwater image enhancement unsupervised learning

Thermally Conductive,Healable Glass Fiber Cloth Reinforced Polymer Composite based onβ-Hydroxyester Bonds Crosslinked Epoxy with Improved Heat Resistance

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Chinese Journal of Polymer Science 2024年第5期42卷 643-654,I0009页

作者： fang Chen Xiao-Yan Pang Ze-Ping zhang Min-zhi Rong Ming-Qiu zhang Key Laboratory for Polymeric Composite and Functional Materials of Ministry of Education GD HPPC LabIGCMESchool of ChemistrySun Yat-sen UniversityGuangzhou510275China Guangdong Provincial Laboratory of Chemistry and Fine Chemical Engineering Jieyang Center Jieyang515200China

To simultaneously endow thermal conductivity,high glass transition temperature(Tg)and healing capability to glass fiber/epoxy(GFREP)composite,dynamic crosslinked epoxy resin bearing reversibleβ-hydroxyl ester bonds was reinforced with boron nitride nanosheets modified glass fiber cloth(GFC@BNNSs).The in-plane heat conduction paths were constructed by electrostatic self-assembly of polyacrylic acid treated GFC and polyethyleneimine decorated ***,the GFC@BNNSs were impregnated with the mixture of lower concentration(3-glycidyloxypropyl)trimethoxysilane grafted BN micron sheets,3,4-epoxycyclohexylmethyl 3,4-epoxycyclohexanecarboxylate and hexahydro-4-methylphthalic anhydride,which accounted for establishing the through-plane heat transport pathways and avoiding serious deterioration of mechanical *** resultant GFREP composite containing less boron nitride particles(17.6 wt%)exhibited superior in-plane(3.29 W·m^(-1)·K^(-1))and through-plane(1.16 W·m^(-1)·K^(-1))thermal conductivities,as well as high Tg of 204℃(Tg of the unfilled epoxy=177℃).The reversible transesterification reaction enabled closure of interlaminar cracks within the composite,achieving decent healing efficiencies estimated by means of tensile strength(71.2%),electrical breakdown strength(83.6%)and thermal conductivity(69.1%).The present work overcame the disadvantages of conventional thermally conductive composites,and provided an efficient approach to prolong the life span of thermally conductive GFREP laminate for high-temperature resistant integrated circuit application.

关键词： Thermally conductive composites Boron nitride High-temperature resistance β-Hydroxyl ester bond Healing

Fabrication and properties of lightweight zirconia with fine closed porosity

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Journal of Iron and Steel Research International 2024年第6期31卷 1449-1458页

作者： Yi-bo zhang Hua-zhi Gu Lv-ping Fu Ao Huang Mei-je zhang The State Key Laboratory of Refractories and Metallurgy Wuhan University of Science and TechnologyWuhan 430081HubeiChina

Lightweight refractory materials with thermal insulation properties and erosion resistance are advantageous for hightemperature *** zirconia refractories were prepared using starch as a pore-forming agent,basic magnesium carbonate as a stabilizer,and nano-zirconia as an *** effects of the nano-zirconia content on the pore and thermal insulation properties of the lightweight zirconia refractories were investigated based on the porosity,phase composition,microstructure,and thermal ***-zirconia was shown to have a high surface energy,and its addition effectively increased the driving force for sintering,reduced the sintering temperature,and promoted the sintering reaction,thereby reducing the apparent porosity and improving the density of the prepared *** to the superplasticity of nano-zirconia,the surface stress caused plastic deformation between particles,which increased the migration rate of grain boundaries and trapped more gas inside the material before it diffused to the surface,thereby enhancing the closed porosity of the *** presence of closed pores could extend the thermal conduction path,decrease the conduction rate,and hinder the conduction effect to effectively reduce the thermal conductivity of the *** a nano-zirconia content of 0.75 wt.%,the prepared lightweight zirconia had the highest closed porosity and the lowest thermal *** apparent porosity,closed porosity,and total porosity of the material were 2.8%,7.0%,and 9.8%,respectively,and the thermal conductivity at 800℃was 1.37 W m^(-1)k^(-1).

关键词： Sintering Nano-zirconia Closed pore:Thermal conductivity Refractory material