检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Heck, Linus Gelbrecht, M. Schaub, M. Boers, N. Institute for Computing and Information Sciences Radboud University Nijmegen Netherlands Earth System Modelling School of Engineering and Design TU Munich Germany Computational Network Science Dept of Computer Science RWTH Aachen University Germany Department of Computer Science RWTH Aachen University Germany Complexity Science Potsdam Institute for Climate Impact Research Potsdam Germany Department of Mathematics and Statistics University of Exeter Exeter United Kingdom

Latent neural stochastic differential equations (SDEs) have recently emerged as a promising approach for learning generative models from stochastic time series data. However, they systematically underestimate the noise level inherent in such data, limiting their ability to capture stochastic dynamics accurately. We investigate this underestimation in detail and propose a straightforward solution: by including an explicit additional noise regularization in the loss function, we are able to learn a model that accurately captures the diffusion component of the data. We demonstrate our results on a conceptual model system that highlights the improved latent neural SDE’s capability to model stochastic bistable dynamics. © 2024, CC BY-NC-ND.

关键词： Stochastic models

来源：评论

学校读者我要写书评

暂无评论

A Modified Polynomial Chaos Modeling Approach for Uncertainty Quantification

A Modified Polynomial Chaos Modeling Approach for Uncertaint...

引用

International Applied Computational Electromagnetics Society Symposium

作者： Majid Ahadi Dolatsara Ambrish Varma Kumar Keshavan Madhavan Swaminathan School of Electrical and Computer Engineering Center for Co-Design of Chip Package System (C3PS) Georgia Institute of Technology Atlanta Ga Cadence Design Systems Inc. Chelmsford MA

ISBN: (纸本)9781728115184;9780996007887

Uncertainty quantification is a key element for modeling high speed circuits, which is often done with Monte Carlo analysis. However, because this method is computationally expensive, new approaches with higher efficiency have been developed. Many popular methods are based on the surrogate models developed with Polynomial Chaos theory. However, size of these models can be prohibitively large for realistic examples. Hence, this paper provides a novel methodology to use an ensemble of weaker models to improve efficiency. Finally, a numerical example with a DDR4 topology is provided.

关键词： DDR4 design space exploration Polynomial chaos theory Uncertainty quantification

来源：评论

学校读者我要写书评

暂无评论

Wave-front arrival time analysis using wide-area frequency measurements

Wave-front arrival time analysis using wide-area frequency m...

引用

IEEE General Meeting Power& Energy Society

作者： Jing Ma Jinlong Li Zengping Wang Jie Wu Tao Xia Lang Chen Yilu Liu School of Electrical and Electronic Engineering North China Electric Power University Beijing China Bradley Department of Electrical and Computer Engineering Virginia Polytechnic Institute Blacksburg VA USA Key Laboratory of Power System Protection and Dynamic Security Monitoring and Control under Ministry of Education North China Electric Power University Beijing China Zhejiang Electric Power Design Institute Hangzhou China

Due to the influence of the local noise at the distribution level, the frequency curve obtained from the frequency disturbance recorder (FDR) unit is too non-stationary to observe its wave-front directly, which causes some problems in the event location estimation method. In this paper, a morphological filter (MF) is employed to extract the wave-front arrival time and the trend of the frequency curve. With the help of the MF, the accuracy of the event location method is greatly improved.

关键词： Frequency measurement Frequency estimation Power system measurements Network servers Wide area measurements Filters Web server Global Positioning system Frequency synchronization Power system dynamics

来源：评论

学校读者我要写书评

暂无评论

Dynamic property evaluations of full-scale 4-story steel frame using parallel FE-analysis

Dynamic property evaluations of full-scale 4-story steel fra...

引用

9th US National and 10th Canadian Conference on Earthquake engineering 2010, Including Papers from the 4th International Tsunami Symposium

作者： Kohiyama, M. Ohsaki, M. Miyamura, T. Onda, K. Isobe, D. Akiba, H. Hori, M. Kajiwara, K. Ine, T. Dept. of System Design Eng. Keio University Yokohama 223-8522 Japan Dept. of Architecture and Architectural Eng. Kyoto University Kyoto 615-8540 Japan Dept. of Computer Science College of Eng. Nihon University Koriyama 963-8642 Japan Allied Engineering Corporation Minato-ku Tokyo 105-0014 Japan Graduate School of Systems and Information Eng. University of Tsukuba Tsukuba 305-8573 Japan Earthquake Research Institute University of Tokyo Bunkyo-ku Tokyo 113-0032 Japan Hyogo Earthquake Engineering Research Center National Research Institute for Earth Science and Disaster Prevention Miki 673-0515 Japan

ISBN: (纸本)9781617388446

The project of E-Simulator is under way at Hyogo Earthquake engineering Research Center (E-Defense) of National Research Institute of Earth Science and Disaster Prevention (NIED), Japan, which facilitates the world's largest shaking table. The E-Simulator uses the parallel finite element (FE) analysis software package called ADVENTYRECluster (ADVC) as a platform. In this study, we report the results of high-precision FE-analysis for simulation of collapse behavior of the 4-story steel building frame that is the specimen of the full-scale total collapse shaking-table test conducted in September 2007 at E-Defense. It is shown that elastoplastic dynamic responses can be estimated with good accuracy using a high-precision FE-analysis without resort to macro models such as plastic hinge and composite beam effect.

关键词： Finite element method

来源：评论

学校读者我要写书评

暂无评论

Mixture of Experts for Network Optimization: A Large Language Model-enabled Approach

arXiv

引用

arXiv 2024年

作者： Du, Hongyang Liu, Guangyuan Lin, Yijing Niyato, Dusit Kang, Jiawen Xiong, Zehui Kim, Dong In The School of Computer Science and Engineering Nanyang Technological University Singapore639798 Singapore The State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications Beijing100876 China The School of Automation Guangdong University of Technology Key Laboratory of Intelligent Information Processing and System Integration of IoT Ministry of Education Guangzhou510006 China Guangdong-HongKong-Macao Joint Laboratory for Smart Discrete Manufacturing Guangzhou510006 China The Pillar of Information Systems Technology and Design Singapore University of Technology and Design Singapore487372 Singapore The Department of Electrical and Computer Engineering Sungkyunkwan University Suwon16419 Korea Republic of

Optimizing various wireless user tasks poses a significant challenge for networking systems because of the expanding range of user requirements. Despite advancements in Deep Reinforcement Learning (DRL), the need for customized optimization tasks for individual users complicates developing and applying numerous DRL models, leading to substantial computation resource and energy consumption and can lead to inconsistent outcomes. To address this issue, we propose a novel approach utilizing a Mixture of Experts (MoE) framework, augmented with Large Language Models (LLMs), to analyze user objectives and constraints effectively, select specialized DRL experts, and weigh each decision from the participating experts. Specifically, we develop a gate network to oversee the expert models, allowing a collective of experts to tackle a wide array of new tasks. Furthermore, we innovatively substitute the traditional gate network with an LLM, leveraging its advanced reasoning capabilities to manage expert model selection for joint decisions. Our proposed method reduces the need to train new DRL models for each unique optimization problem, decreasing energy consumption and AI model implementation costs. The LLM-enabled MoE approach is validated through a general maze navigation task and a specific network service provider utility maximization task, demonstrating its effectiveness and practical applicability in optimizing complex networking systems. © 2024, CC BY.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Building Energy Efficiency Assessment Base on Predict-Center Criterion Under Diversified Conditions

SSRN

引用

SSRN 2023年

作者： Liao, Xuechao Zheng, Xiujuan Ping, Zouwei Zhao, Haoyi Wang, Ning Zhang, Yong School of Computer Science and Technology Wuhan University of Science and Technology Wuhan430080 China Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System Wuhan430080 China School of Information Science and Engineering Wuhan University of Science and Technology Wuhan430080 China National Key Laboratory of Science and Technology on Vessel Integrated Power System Naval University of Engineering Wuhan430030 China WISDRI City Construction Engineering &Research Incorporation Ltd. Wuhan430030 China Changjiang Institute of Survey Planning Design and Research Wuhan430010 China

With the rapid expanding of urban economy and population, how to reduce the energy consumption and improve energy efficiency of public buildings has become a key issue for energy conservation and urban development. However, due to the diversity of building employing condition, energy consumption data often has problems with nonlinearity, outliers, and diversity, which will greatly reduce the accuracy of energy consumption prediction and energy conservation assessment. To overcome the above drawbacks, a regression prediction model of building energy consumption is designed, which cluster the same type of data characteristics to form a subset of equipment through Gaussian mixture clustering, and design the specific normalized adaptive loss function to determine the optimal control parameters of regression model. Furthermore, a novel predict-center energy efficiency evaluation criterion is proposed, and the cumulative distribution pattern is further statistically analyzed. Finally, the energy benchmark evaluation framework is proposed to determine the energy efficiency evaluation grade. Through comparative experimental analysis, it can be seen that the proposed prediction model can adapt to different types of building data, and have better prediction accuracy and convergence speed. The consistency and robustness of the benchmark evaluation framework and method are optimal. © 2023, The Authors. All rights reserved.

关键词： Energy utilization

来源：评论

学校读者我要写书评

暂无评论

Vortex detection on unsteady CFD simulations using recurrent neural networks 48th

Vortex detection on unsteady CFD simulations using recurrent...

引用

48th AIAA Fluid Dynamics Conference, 2018

作者： Rajendran, Vinod Ke, Yiping Leonardi, Erwin Menzies, Kevin Nanyang Technological University 50 Nanyang Avenue Singapore639798 Singapore School of Computer Science and Engineering Nanyang Technological University 50 Nanyang Avenue Singapore639798 Singapore Rolls-Royce Singapore Pte Ltd 1 Seletar Aerospace Crescent Singapore797565 Singapore Rolls-Royce plc System Design Whittle House WH64 P O Box 3 Filton BristolBS34 7QE United Kingdom Applied Technology Group Rolls-Royce Singapore Pte Ltd 1 Seletar Aerospace Crescent 797565 Singapore

ISBN: (数字)9781624105531

ISBN: (纸本)9781624105531

The formation of vortices in the surrounding fluid is used as one of the indicators for measuring the efficiency of aerodynamic system designs. Automated vortex detection methods are needed to facilitate the exploration of large scale turbulent fluid flow simulations. Following the success of deep learning applications and algorithms, this paper details a data driven approach for detecting vortices using recurrent neural networks (RNNs). Long Short Term Memory (LSTM) and Gated Recurrent Unit (GRU) are the RNN architectures utilized in this paper due to their ability to support long term memory. The method is able to identify if a data point belongs a vortex or not in different time frames by simply using its pressure, density and velocity attributes. The efficacy of the proposed approach is demonstrated on a 3D case of an unsteady flow around a cylinder. The experimental results indicates that the RNNs can detect vortices effectively without relying on feature engineering like in traditional machine learning methods. Furthermore, it is able to obtain improved performance in identifying vortices and capturing hidden patterns due to their ability to remember information over time. The prime advantage of this approach is that the vortices can be predicted rapidly. Hence, RNN has the potential to be used as a predictive tool in vortex detection. © 2018, American Institute of Aeronautics and Astronautics Inc, AIAA. All rights reserved.

关键词： Vortex flow

来源：评论

学校读者我要写书评

暂无评论

Jigsaw-based Benchmarking for Learning Robotic Manipulation

arXiv

引用

arXiv 2023年

作者： Liu, Xiaobo Wan, Fang Ge, Sheng Wang, Haokun Sun, Haoran Song, Chaoyang Department of Mechanical and Energy Engineering Southern University of Science and Technology Guangdong Shenzhen518055 China School of Design Southern University of Science and Technology Guangdong Shenzhen518055 China Robotics & Autonomous Systems Thrust System Hub Hong Kong University of Science and Technology Hong Kong Department of Computer Science The University of Hong Kong Hong Kong

Benchmarking provides experimental evidence of the scientific baseline to enhance the progression of fundamental research, which is also applicable to robotics. In this paper, we propose a method to benchmark metrics of robotic manipulation, which addresses the spatial-temporal reasoning skills for robot learning with the jigsaw game. In particular, our approach exploits a simple set of jigsaw pieces by designing a structured protocol, which can be highly customizable according to a wide range of task specifications. Researchers can selectively adopt the proposed protocol to benchmark their research outputs, on a comparable scale in the functional, task, and system-level of details. The purpose is to provide a potential look-up table for learning-based robot manipulation, commonly available in other engineering disciplines, to facilitate the adoption of robotics through calculated, empirical, and systematic experimental evidence. © 2023, CC BY.

关键词： Benchmarking

来源：评论

学校读者我要写书评

暂无评论

Pulling Figure of Dielectric Resonator for 3D Filament Defectoscopy

Pulling Figure of Dielectric Resonator for 3D Filament Defec...

引用

International Scientific Symposium on Metrology and Metrology Assurance (MMA)

作者： Maryna Miroshnyk Olga Zaichenko Anatolii Miroshnyk Nataliia Hapon Yurii Pakhomov Liubov Klymenko dept. of Theoretical and Applied System Engineering V. N. Karazin Kharkiv National University Kharkov Ukraine dept. of Design and Operation of Electronic Devices Kharkiv National University of Radioelectronics Kharkiv Ukraine dept. of Design Automation Kharkiv National University of Radioelectronics) Kharkiv Ukraine dept. of Computer Science and Information Technologies O.M. Beketov National University of Urban Economy in Kharkiv Kharkiv Ukraine dept. of Specialized Computer Systems Ukrainian State University of Railway Transport Kharkiv Ukraine

ISBN: (数字)9798350352832

ISBN: (纸本)9798350352849

The research aim is to derive connection between the parameters of dielectric resonators frequency shift and the warp defect of 3D printing filament. Object of investigation are the processes of interaction of electromagnetic wave of a dielectric resonator with filament for 3D printing. The proposed model describes the connection between the frequency shift and defect warp of filament frequency pulling depending on the load.

关键词： Electromagnetic scattering Resonant frequency Estimation Metrology Three-dimensional printing Resonance Dielectrics Frequency measurement Resonators Load modeling

来源：评论

学校读者我要写书评

暂无评论

Missing step detection in a high speed micro stepping motor using current feedback

Missing step detection in a high speed micro stepping motor ...

引用

International Conference on Control, Automation and systems ( ICCAS)

作者： Sung-Wook Moon Young-Jin Kim Yoon-Taek Lim Dong Hwan Kim NID Graduate School Seoul National University of Science and Technology Seoul South Korea Nano Manufacturing Research Center Seoul National University of Science and Technology Seoul South Korea Convex Computer Corporation Seoul South Korea Department of Mechanical System Design Engineering Seoul National University of Science and Technology Seoul South Korea

In this work, new approach of missing step detection without using costly encoder in a micro stepping motor is proposed. For a steeping motor driver, a MOSFET driven H-bridge driver was adopted, and sinusoidal type of current is controlled to implement the micro stepping. Also, in order to investigate the missing step the current flowing the coil which is generated from the operational amplifier was given feedback and the feedback data was analyzed to detect the missing step. As a sequence, the missing step was compensated as much as the amount of the missing steps, yielding the control of the desired micro stepping angle.

关键词： Coils Torque Position control Educational institutions MOSFET circuits Power electronics

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：