检索结果-南通市图书馆

维普期刊数据库

Comparing four machine learning algorithms for household non-intrusive load monitoring

在线全文

学校读者我要写书评

暂无评论

Energy and AI 2024年第3期17卷 206-219页

作者： Thomas lee Young james Gopsill Maria Valero Sindre Eikevåg Ben Hicks University of Bristol UK

The combination of Machine Learning (ML), smart energy meters, and availability of household appliance energy profile data has opened new opportunities for Non-Intrusive Load Monitoring (NILM). However, the number of options makes it challenging in selecting optimal combinations for different energy applications, which requires studies to examine their *** paper contributes one such study that investigated four established ML approaches – K Nearest Neighbour (KNN), Support Vector Machine (SVM), eXtreme Gradient Boosting (XGBoost) and Convolutional Neural Network (CNN) – and their performance in classifying appliance events from Alternating Current (AC) and Root Mean Square (RMS) energy data where the sampling frequency and training dataset set size was varied (10 Hz–1 kHz and 50–2000 examples per class, respectively). The computational expense during training, testing and storage was also assessed and evaluated with reference to real-world *** CNN classifier trained on AC data at 500 Hz and 11,000 examples gave the best F1-score (0.989) followed by the KNN classifier (0.940). The storage size required by the CNN models was 3̃MB, which is very close to fitting on cost-effective embedded system microcontrollers. This would prevent high-rate data needing to be sent to the cloud as analysis could be performed on edge computing Internet-of-Things (IoT) devices.

关键词： NILM ML Edge computing Signal processing Smart meters

Time-Aware LSTM Neural Networks for Dynamic Personalized Recommendation on Business Intelligence

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Tsinghua Science and Technology 2024年第1期29卷 185-196页

作者： Xuan Yang james A.Esquivel Graduate School Angeles University FoundationAngeles City 2009Philippines Shandong Provincial University Laboratory for Protected Horticulture Weifang University of Science and TechnologyWeifang 262700China

Personalized recommendation plays a critical role in providing decision-making support for product and service analysis in the field of business ***,deep neural network-based sequential recommendation models gained considerable ***,existing approaches pay litle attention to users'dynamically evolving interests,which are influenced by product attributes,especially product *** overcome these challenges,we propose a dynamic personalized recommendation model:***,we first embed product information and attribute information into a unified data ***,we exploit long short-term memory(LsTM)networks to characterize sequential behavior over multiple time periods and seize evolving interests by hierarchical LSTM ***,similarity values between users are measured through pairwise interest features,and personalized recommendation lists are generated.A series of experiments reveal the superiority of the proposed method compared withotheradvanced methods.

关键词： personalized recommendations evolving interests embedding LsTM networks

Probing turbulence in galaxies with fast radio burst scintillation

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) 2024年第1期67卷 177-178页

作者： james M.Cordes Department of Astronomy the Cornell Center for Astrophysics and Planetary ScienceCornell UniversityIthaca NY 14853USA

Interstellar scintillation(ISS)is a well-known phenomenon observed from pulsars in the Milky *** comprises intensity variations *** and frequency caused by constructive and destructive interference from multipath propagation,which is due to diffraction from fluctuations in electron density in the interstellar medium(ISM)on scales as small as~100 *** density fluctuations appear to be part of a turbulence process combined with discrete structures in electron density from other astrophysical processes;together these fluctuations extend to scales of pc or larger.

关键词： scales galaxies turbulence

维普期刊数据库

LSTM Network-Based Adaptation Approach for Dynamic Integration in Intelligent End-Edge-Cloud Systems

在线全文

学校读者我要写书评

暂无评论

Tsinghua Science and Technology 2024年第4期29卷 1219-1231页

作者： Xuan Yang james A.Esquivel Graduate School Angeles University FoundationAngeles City 2009Philippinesand also with Shandong Provincial University Laboratory for Protected HorticultureWeifang University of Science and TechnologyWeifang 262700China Graduate School Angeles University FoundationAngeles City 2009Philippines

Edge computing, which migrates compute-intensive tasks to run on the storage resources of edge devices, efficiently reduces data transmission loss and protects data privacy. However, due to limited computing resources and storage capacity, edge devices fail to support real-time streaming data query and processing. To address this challenge, first, we propose a Long Short-Term Memory (LSTM) network-based adaptive approach in the intelligent end-edge-cloud system. Specifically, we maximize the Quality of Experience (QoE) of users by automatically adapting their resource requirements to the storage capacity of edge devices through an event mechanism. Second, to reduce the uncertainty and non-complete adaption of the edge device towards the user’s requirements, we use the LSTM network to analyze the storage capacity of the edge device in real time. Finally, the storage features of the edge devices are aggregated to the cloud to re-evaluate the comprehensive capability of the edge devices and ensure the fast response of the user devices during the dynamic adaptation matching process. A series of experimental results show that the proposed approach has superior performance compared with traditional centralized and matrix decomposition based approaches.

关键词： quality of experience data query end-edge-cloud Long Short-Term Memory(LSTM)networks

Analysis of the Oscillatory Flows of Multiple Shock Waves in a Constant Area Duct

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 Thermal Science 2024年第3期33卷 794-806页

作者： K james Jintu KIM Heuy Dong Department of Mechanical Engineering Andong National UniversityAndong36729Republic of Korea CAD-IT Korea Co. Ltd.Seoul08591Republic of Korea

The oscillatory response of multiple shock waves to downstream perturbations in a supersonic flow is studied numerically in a rectangular *** shock waves are formed inside the duct at a shock Mach number of *** duct has an exit height of H,and the effect of duct resonance on multiple shock oscillations is investigated by attaching exit ducts of lengths 0H,50H,and *** downstream disturbance frequency varied from 10 Hz to 200 Hz to explore the oscillation characteristics of the multiple shock *** oscillatory response of shock waves under self-excited and forced oscillation conditions are analyzed in terms of wall static pressure,shock train leading-edge location,shock train length,and the size of the separation *** extent of the initial shock location increases with an increase in exit duct length for the self-excited oscillation *** analysis of the shock train leading edge and the spectral analysis of wall static pressure variations are *** variation in the shock train length is analyzed using the pressure ratio method for self-excited as well as forced *** RMS amplitude of the normalized shock train length(ζ_(ST))increases with an increase in the exit duct length for the self-excited oscillation *** the downstream perturbation frequency is increased,ζ_(ST)is decreased for exit duct *** all exit duct designs and downstream forcing frequencies,the size of the separation bubble grows and shrinks during the shock oscillations,demonstrating the dependence on duct resonance and forced oscillations.

关键词： shock train downstream disturbance supersonic flow shock wave boundary layer interaction duct resonance flow separation

维普期刊数据库

Learning physically based material and lighting decompositions for face editing

在线全文

学校读者我要写书评

暂无评论

Computational Visual Media 2024年第2期10卷 295-308页

作者： Qian Zhang Vikas Thamizharasan james Tompkin Brown University ProvidenceRhode Island 02906USA

Lighting is crucial for portrait photography,yet the complex interactions between the skin and incident light are expensive to model computationally in graphics and difficult to reconstruct analytically via computer ***,to allow fast and controllable reflectance and lighting editing,we developed a physically based decomposition through deep learned priors from path-traced portrait *** approaches that used simplified material models or low-frequency or low-dynamic-range lighting struggled to model specular reflections or relight directly without intermediate ***,we estimate the surface normal,skin albedo and roughness,and high-frequency HDRI maps,and propose an architecture to estimate both diffuse and specular reflectance *** our experiments,we show that this approach can represent the true appearance function more effectively than simpler baseline methods,leading to better generalization and higher-quality editing.

关键词： intrinsic decomposition portrait relighting inverse rendering deep learning

Event-triggered sliding mode control of linear repetitive processes and its application in metal rolling process

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(Information Sciences) 2024年第3期67卷 167-181页

作者： Xinyu LV Yugang NIU james LAM Key Laboratory of Smart Manufacturing in Energy Chemical Process (East China University of Science and Technology) Ministry of Education Department of Mechanical Engineering The University of Hong Kong

This paper investigates the design problem of a sliding mode controller for linear repetitive processes(LRPs) with a finite pass length on each pass. Under limited communication resources, an eventtriggered mechanism is implemented from the sensors to the controller, whose triggered sequence is consistent with the evolution direction of LRPs. To periodically orchestrate communication between the controller and actuator, a round-robin scheduling mechanism is established based on the pass length. A sliding function involving the state and pass profile is constructed, and the sliding mode controller is designed using triggered states and pass profile signals. Feasible controller gains are attained to ensure the stability of the resultant closed-loop system. Finally, the proposed strategy is implemented for a multi-roll metal rolling process.

关键词： event-triggered mechanism linear repetitive processes metal rolling process round-robin scheduling sliding mode control

同方期刊数据库

Genetic risk stratification of inflammatory bowel disease-associated venous thromboembolism:An Asian perspective

在线全文

学校读者我要写书评

暂无评论

World Journal of Gastroenterology 2024年第9期30卷 1250-1252页

作者： james Guoxian Huang Division of Gastroenterology Hepatology and NutritionDepartment of PaediatricsKhoo Teck Puat-National University Children’s Medical InstituteNational University Health SystemSingapore 119228Singapore Department of Paediatrics Yong Loo Lin School of MedicineNational University of SingaporeSingapore 119228Singapore

The utilisation of polygenic scoring models may enhance the clinician’s ability to risk stratify an inflammatory bowel disease patient’s individual risk for venous thromboembolism(VTE)and guide the appropriate usage... 详细信息

关键词： Thromboembolism Inflammatory bowel disease Genetic screening Venous thromboembolism Thromboprophylaxis

An analytical model to estimate the time delay to reach spontaneous ignition considering heat loss in oil reservoirs

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Petroleum Science 2024年第4期21卷 2469-2474页

作者： james J.Sheng Er-Long Yang College of Petroleum Engineering Northeast Petroleum University

During air injection into an oil reservoir, an oxidation reaction generates some heat to raise the reservoir temperature. When the reservoir temperature reaches an ignition temperature, spontaneous ignition occurs. There is a time delay from the injection to ignition. There are mixed results regarding the feasibility of spontaneous ignition in real-field projects and in laboratory experiments. No analytical model is available in the literature to estimate the oxidation time required to reach spontaneous ignition with heat *** paper discusses the feasibility of spontaneous ignition from theoretical points and experimental and field project observations. An analytical model considering heat loss is proposed. Analytical models with and without heat loss investigate the factors that affect spontaneous ignition. Based on the discussion and investigations, we find that it is more difficult for spontaneous ignition to occur in laboratory experiments than in oil reservoirs; spontaneous ignition is strongly affected by the initial reservoir temperature, oil activity, and heat loss; spontaneous ignition is only possible when the initial reservoir temperature is high, the oil oxidation rate is high, and the heat loss is low.

关键词： Spontaneous ignition Time delay Air injection Heat loss Low temperature oxidation Combustion