检索结果-南通市图书馆

基于上下文生成对抗网络的时间序列异常检测方法

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

哈尔滨工业大学学报 2024年第5期56卷 1-11页

作者：胡智超余翔湛刘立坤张宇于海宁哈尔滨工业大学网络空间安全学院哈尔滨150001

时间序列的异常检测是网络服务保障、数据安全检测、系统监控分析等应用中所依赖的一项关键技术。为解决在实际场景的时间序列异常检测中由于时间序列上下文的模糊性、数据分布的复杂性以及异常检测模型的不确定性所带来的异常检测结果... 详细信息

时间序列的异常检测是网络服务保障、数据安全检测、系统监控分析等应用中所依赖的一项关键技术。为解决在实际场景的时间序列异常检测中由于时间序列上下文的模糊性、数据分布的复杂性以及异常检测模型的不确定性所带来的异常检测结果的有效性、合理性、稳定性等不足的问题,本文提出了一种新的基于上下文生成对抗网络的时间序列异常检测方法AdcGAN。首先,通过处理历史数据,提取用于生成时序数据的条件上下文;然后,采用条件生成对抗网络的设计策略,使用条件上下文,构建上下文生成对抗网络,实现对任意时刻数据的条件分布预测,同时AdcGAN采用Dropout近似模型不确定性,使用概率分布代替点估计作为预测结果;接着,从观测的差异(用期望偏差表示)和模型的不确定性(用预测方差表示)两个方面来衡量异常;最后,提出基于数据统计信息的异常阈值自动设置方法,减少手动调节的参数量。实验结果表明,与同类基准算法进行对比,在NAB数据集中的47个真实时序数据上,本文提出的AdcGAN可以有效地检测出时序数据中的异常,在大多数评价指标上都优于其他基准方法,并且具有更好的稳定性。

关键词：时间序列异常检测生成对抗网络模型不确定性生成模型深度学习

维普期刊数据库博看期刊

Characteristics of a pre-combustion plasma jet igniter

在线全文

学校读者我要写书评

暂无评论

Chinese Journal of Aeronautics 2024年第7期37卷 178-189页

作者： Jinlu yu Baowen ZHANG Yang yu Bingbing ZHAO Lei ZHANG Aviation Engineering School Air Force Engineering UniversityXi’an 710038China Equipment Management and Unmanned Aerial Vehicle Engineering School Air Force Engineering UniversityXi’an 710038China

Plasma ignition technology has delivered good performance in the aerospace industry. In this study, a pre-combustion plasma jet igniter was designed, and its characteristics were examined from three aspects: the morphology, temperature, and discharge characteristics and process of ignition. Images of the OH distribution were obtained by using an OH Planar Laser-Induced Fluorescence(OH-PLIF) experimental system. Results have shown that the proposed plasma jet had a higher OH concentration, longer length, and larger area than those of a traditional igniter. The stability of discharge of the igniter was improved as the equivalence ratio φ was increased, and reducing gas flow reduced the pulsation of the plasma jet. When the input current was increased from 15A to 35 A, the highest average temperature increased from 5127 K to 7987 K. An increase in the equivalence ratio reduced the region of arc ionization, but expanded the regions of the core combustion reaction and the outer flame. Herein, this study has obtained a deep understanding of the jet and ignition law and developed a new idea for the application of plasma in the ignition field.A pre-combustion plasma jet igniter can significantly improve the efficiency of ignition and shorten the ignition process compared with a traditional igniter.

关键词： Pre-combustion plasma igniter Plasma Jet temperature field Ignition process Arc Flame

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

在线全文

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

Macroporous Directed and Interconnected Carbon Architectures Endow Amorphous Silicon Nanodots as Low‑Strain and Fast‑Charging Anode for Lithium‑Ion Batteries

维普期刊数据库评论

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5期16卷 333-351页

作者： Zhenwei Li Meisheng Han Peilun yu Junsheng Lin Jie yu Guangdong Provincial Key Laboratory of Semiconductor Optoelectronic Materials and Intelligent Photonic Systems Shenzhen Engineering Lab for Supercapacitor MaterialsSchool of Material Science and EngineeringHarbin Institute of TechnologyShenzhenUniversity TownShenzhen 518055People’s Republic of China Songshan Lake Materials Laboratory Dongguan Dongguan 523808GuangdongPeople’s Republic of China Department of Mechanical and Energy Engineering Southern University of Science and TechnologyShenzhen 518055People’s Republic of China

Fabricating low-strain and fast-charging silicon-carbon composite anodes is highly desired but remains a huge challenge for lithium-ion ***,we report a unique silicon-carbon composite fabricated by uniformly dis-persing amorphous Si nanodots(SiNDs)in carbon nanospheres(SiNDs/C)that are welded on the wall of the macroporous carbon framework(MPCF)by vertical graphene(VG),labeled as MPCF@VG@SiNDs/*** high dispersity and amor-phous features of ultrasmall SiNDs(~0.7 nm),the flexible and directed electron/Li+transport channels of VG,and the MPCF impart the MPCF@VG@SiNDs/C more lithium storage sites,rapid Li+transport path,and unique low-strain property during Li+***,the MPCF@VG@SiNDs/C exhibits high cycle stability(1301.4 mAh g^(-1) at 1 A g^(-1) after 1000 cycles without apparent decay)and high rate capacity(910.3 mAh g^(-1),20 A g^(-1))in half cells based on industrial electrode *** assembled pouch full cell delivers a high energy density(1694.0 Wh L^(-1);602.8 Wh kg^(-1))and an excellent fast-charging capability(498.5 Wh kg^(-1),charging for 16.8 min at 3 C).This study opens new possibilities for preparing advanced silicon-carbon com-posite anodes for practical applications.

关键词： Amorphous Si nanodots Low-strain Fast-charging Lithium-ion batteries

Hybrid pedestrian positioning system using wearable inertial sensors and ultrasonic ranging

在线全文

学校读者我要写书评

暂无评论

Defence Technology（防务技术） 2024年第3期33卷 327-338页

作者： Lin Qi yu Liu Chuanshun Gao Tao Feng yue yu School of Communication and Information Engineering Chongqing University of Posts and TelecommunicationsChongqing 400065China Department of Land Surveying and Geo-Informatics The Hong Kong Polytechnic UniversityHong Kong 999077China

Pedestrian positioning system(PPS)using wearable inertial sensors has wide applications towards various emerging fields such as smart healthcare,emergency rescue,soldier positioning,*** performance of traditional PPS is limited by the cumulative error of inertial sensors,complex motion modes of pedestrians,and the low robustness of the multi-sensor collaboration *** paper presents a hybrid pedestrian positioning system using the combination of wearable inertial sensors and ultrasonic ranging(H-PPS).A robust two nodes integration structure is developed to adaptively combine the motion data acquired from the single waist-mounted and foot-mounted node,and enhanced by a novel ellipsoid constraint *** addition,a deep-learning-based walking speed estimator is proposed by considering all the motion features provided by different nodes,which effectively reduces the cumulative error originating from inertial ***,a comprehensive data and model dual-driven model is presented to effectively combine the motion data provided by different sensor nodes and walking speed estimator,and multi-level constraints are extracted to further improve the performance of the overall *** results indicate that the proposed H-PPS significantly improves the performance of the single PPS and outperforms existing algorithms in accuracy index under complex indoor scenarios.

关键词： Pedestrian positioning system Wearable inertial sensors Ultrasonic ranging Deep-learning Data and model dual-driven

Ignition processes and characteristics of charring conductive polymers with a cavity geometry in precombustion chamber for applications in micro/nano satellite hybrid rocket motors

在线全文

学校读者我要写书评

暂无评论

Defence Technology（防务技术） 2024年第2期32卷 55-66页

作者： Zhiyuan Zhang Hanyu Deng Wenhe Liao Bin yu Zai yu Nanjing University of Science and Technology Nanjing 210094China

The arc ignition system based on charring polymers has advantages of simple structure,low ignition power consumption and multiple ignitions,which bringing it broadly application prospect in hybrid propulsion system of micro/nano ***,charring polymers alone need a relatively high input voltage to achieve pyrolysis and ignition,which increases the burden and cost of the power system of micro/nano satellite in practical *** conductive substance into charring polymers can effectively decrease the conducting voltage which can realize low voltage and low power consumption repeated ignition of arc ignition *** this paper,a charring conductive polymer ignition grain with a cavity geometry in precombustion chamber,which is composed of PLA and multiwall carbon nanotubes(MWCNT)was *** detailed ignition processes were analyzed and two different ignition mechanisms in the cavity of charring conductive polymers were *** ignition characteristics of charring conductive polymers were also investigated at different input voltages,ignition grain structures,ignition locations and injection schemes in a visual ignition *** results demonstrated that the ignition delay and external energy required for ignition were inversely correlated with the voltages applied to ignition ***,the incremental depth of cavity shortened the ignition delay and external energy required for ignition while accelerated the propagation of *** the depth of cavity increased from 2 to 6 mm(at 50 V),the time of flame propagating out of ignition grain changed from 235.6 to 108 ms,and values of mean ignition delay time and mean external energy required for ignition decreased from 462.8 to 320 ms and 16.2 to 10.75 J,*** rear side of the cavity was the ideal ignition position which had a shorter ignition delay and a faster flame propagation speed in comparison to other ignition *** to direct injection scheme,swirlin

关键词： Micro/nano satellite hybrid propulsion Arc ignition Charring conductive polymer Ignition mechanism Ignition characteristic Repeated ignition

Chord length sampling correction analysis for dispersion fuel in Monte Carlo simulation

在线全文

学校读者我要写书评

暂无评论

Nuclear Science and Techniques 2024年第4期35卷 57-64页

作者： Zhao-yu Liang Ding She yu-Tong Wen Lei Shi Institute of Nuclear and New Energy Technology(INET) Key Laboratory of Advanced Reactor Engineering and Safety of Ministry of EducationTsinghua UniversityBeijing 100084China

Dispersion fuels,knowned for their excellent safety performance,are widely used in advanced reactors,such as hightemperature gas-cooled *** with deterministic methods,the Monte Carlo method has more advantages in the geometric modeling of stochastic *** explicit modeling method has high computational accuracy and high computational *** chord length sampling(CLS)method can improve computational efficiency by sampling the chord length during neutron transport using the matrix chord length?s probability density *** study shows that the excluded-volume effect in realistic stochastic media can introduce certain deviations into the CLS.A chord length correction approach is proposed to obtain the chord length correction factor by developing the Particle code based on equivalent transmission *** numerical analysis against reference solutions from explicit modeling in the RMC code,it was demonstrated that CLS with the proposed correction method provides good accuracy for addressing the excludedvolume effect in realistic infinite stochastic media.

关键词： Stochastic media Monte Carlo Chord length sampling Excluded-volume effect Chord length correction

Heat Transfer Characteristics of a Microchannel Heat Sink with Highly-Dense Micro-Jet Arrays

在线全文

学校读者我要写书评

暂无评论

Journal of Thermal Science 2024年第3期33卷 1148-1160页

作者： ZHANG yu CHEN Xiaoyan MIAO Lin CHEN Liang HOU yu State Key Laboratory of Multiphase Flow in Power Engineering Xi’an Jiaotong UniversityXi’an710049China Beijing Aerospace Technology Institute Beijing100083China

This paper proposed a new structure of highly-dense micro-jet arrays for hybrid jet-impingement/microchannel heat sinks(~120 jet per cm^(2) with jet width of 0.35 mm).Parametric study is performed to investigate the influence of structure and flow parameters on the convective heat transfer of water jet cooling in confined *** simulation results show that the optimal jet width,jet spacing and impingement distance are around 0.2 mm,0.2 mm and 0.3 mm,respectively,which can achieve a low thermal resistance as well as a relatively low pressure *** analysis of the heat transfer pathways shows that the micro-fin jet structures can extend the impinging heat transfer area and conduct a considerable proportion of heat,which can reach up to 38.9%with the Reynolds number ranging from 797.1 to *** heat transfer characteristics in the heat sink will shift from impingement dominated heat transfer to the channel-convection dominated heat transfer as the jet impingement distance increases.A correlation is proposed to predict the average Nusselt number on the stagnation area for the heat sink with different structure parameters,and the deviations of predictions from the numerical results are less than 10%.

关键词： hybrid heat sink impingement microchannel thermal management high heat flux

Research on the mechanism of the two-dimensional ultrasonic surface burnishing process to enhance the wear resistance for aluminum alloy

在线全文

学校读者我要写书评

暂无评论

Friction 2024年第3期12卷 490-509页

作者： Zhen-yu ZHOU Qiu-Yang ZHENG yu LI Cong DING Guang-Jian PENG Zhong-yu PIAO College of Mechanical Engineering Zhejiang University of TechnologyHangzhou 310023China Key Laboratory of Special Purpose Equipment and Advanced Processing Technology Ministry of Education and Zhejiang ProvinceZhejiang University of TechnologyHangzhou 310014China

The gradient nanostructure is machined on the aluminum(Al)alloy by the two-dimensional ultrasonic surface burnishing process(2D-USBP).The mechanism of why the gradient nanostructure enhances wear resistance is *** mechanical properties and microstructure characterization for the gradient nanostructure are performed by operating a nanoindenter,transmission electron microscopy(TEM),and electron backscattered diffraction(EBSD).Dry wear tests are performed on the samples before and after machining to evaluate the wear resistance and *** effect of the gradient nanostructure on the wear resistance is explored by developing the crystal plasticity(CP)finite element and molecular dynamics(MD)*** characterization results show that the 2D-USBP sample prepared a gradient structure of~600μm thick on the aluminum surface,increasing the surface hardness from 1.13 to 1.71 GPa and reducing the elastic modulus from 78.84 to 70.14 *** optimization of the surface microstructure and the increase of the mechanical properties effectively enhance the wear resistance of the sample,with 41.20%,39.07%,and 54.58% of the wear scar areas for the 2D-USBP treated samples to the original samples under 5,10,and 15 N loads,*** gradient nanostructure hinders the slip of dislocations inside the sample during the wear process and reduces the size and scope of plastic deformation;meanwhile,the resistance to deformation,adhesion,and crack initiation and propagation of the sample surface is improved,resulting in enhanced wear resistance.

关键词： structural superlubricity lattice registry elastic deformation strained solitons crystal plasticity molecular dynamics simulations