检索结果-内蒙古大学图书馆

作者： Nagpal, Arun Zhou, Ming Ilic, Ognjen Yu, Zongfu Atwater, Harry A. Thomas J. Watson Laboratory of Applied Physics California Institute of Technology 1200 E. California Avenue PasadenaCA91125 United States Department of Electrical Engineering Stanford University 450 Serra Mall StanfordCA94305 United States Department of Mechanical Engineering University of Minnesota 111 Church Street SE MinneapolisMN55455 United States Department of Electrical and Computer Engineering University of Wisconsin 1415 Engineering Drive MadisonWI53706 United States

We report the design of a tunable, narrowband, thermal metasurface that employs a hybrid resonance generated by coupling a tunable permittivity graphene ribbon to a silicon photonic crystal. The gated graphene ribbon array, proximitized to a high quality factor Si photonic crystal supporting a guided mode resonance, exhibits tunable narrowband absorbance lineshapes (Q > 10,000). Actively tuned Fermi level modulation in graphene with applied gate voltage between high absorptivity and low absorptivity states gives rise to absorbance on/off ratios exceeding 60. We employ coupled-mode theory as a computationally efficient approach to elements of the metasurface design, demonstrating an orders of magnitude speedup over typical finite element computational methods. © 2023 Optica Publishing Group.

关键词： Absorbance

来源：评论

学校读者我要写书评

暂无评论

Evaluating the thermal performance of perovskite SrSnO3 field effect transistors

Evaluating the thermal performance of perovskite SrSnO3 fiel...

引用

Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITHERM)

作者： Bivek Bista Prafful Golani Fengdeng Liu Tristan Truttmann Georges Pavlidis Andrea Centrone Bharat Jalan Steven Koester Department of Mechanical Engineering University of Connecticut Storrs Connecticut USA Department of Electrical and Computer Engineering University of Minnesota Minneapolis Minnesota USA Department of Chemical Engineering and Materials Science University of Minnesota Minneapolis Minnesota USA Physical Measurement Laboratory National Institute of Standards and Technology Gaithersburg Maryland USA

Given the ever increasing, global electricity consumption, improving the efficiency and reliability of high-power electronics is of paramount importance. Ultra-wide band gap (> 3.4 eV) semiconductors have shown the potential to be used in the next generation of power electronics due to their high breakdown field and mobility. Specifically, high-quality growth of doped strontium stannate perovskite-oxide (SSO) has been recently demonstrated. The thermal properties of this novel material, however, have not been fully investigated and could be the limiting factor (excessive junction temperatures) to maximize device performance. In this study, high resolution $(\approx \mathbf{410}\ \mathbf{nm})$ transient thermoreflectance imaging (TTI) is used to obtain temperature maps of SSO two-terminal devices with varying channel width ( $\mathbf{10}\ \boldsymbol{\mu} \mathbf{m}$ to $\mathbf{20}\ \boldsymbol{\mu} \mathbf{m}$ ). The results show that the device thermal resistance increases by $\approx \mathbf{21} {\%}$ due to the increase in thermal spreading resistance with channel width. The observed trend is further confirmed via transient thermal analysis where the thermal time constant is shown to increase from $\mathbf{20}.\mathbf{9}\ \boldsymbol{\mu}\mathbf{s}$ to $\mathbf{29}.\mathbf{9}\ \boldsymbol{\mu}\mathbf{s}$ . Overall, the active layer thicknesses and device geometry (length and width), must be carefully considered to improve device performance and lifetime.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Demo: Intelligent Abnormal Detection for electrical Equipment Based on Machine Learning

Demo: Intelligent Abnormal Detection for Electrical Equipmen...

引用

IEEE Internatonal Conference on Mobile Adhoc and Sensor Systems (MASS)

作者： Jia-Ming Liang Shashank Mishra You-Jian Ma Ping-Jui Yang Ming-Ren Ou Chun-Wen Ku Jyun-You Huang Jen-Jee Chen Department of Electrical Engineering National University of Tainan Tainan Taiwan Department of Computer Science and Engineering University of Minnesota Minneapolis USA Institute of Computational Intelligence National Yang Ming Chiao Tung University Tainan Taiwan

ISBN: (数字)9798350363999

ISBN: (纸本)9798350364002

electrical equipment is essential for factories, offices, and daily life. Neglecting its proper maintenance can result in fires or hazardous accidents. Currently, traditional electrical plugs can only confirm whether the electrical equipment is overloaded and cannot detect hazards or aging abnormalities in real-time. Therefore, an intelligent abnormal detection system for electrical equipment is required. Such a system needs to detect electrical information of the equipment, including voltage, current, power factor, and frequency, and combine it with AI techniques to dynamically detect abnormal conditions in the electrical equipment. In view of this, we proposed and developed an AIoT system with the following functional components: 1. Smart power sensing devices for collecting and monitoring abnormal events. 2. A real-time anomaly alerting app to receive abnormal notifications. 3. A highly efficient and fast anomaly recognition model, capable of identifying various abnormal conditions in electrical equipment. 4. A visualized electrical equipment analysis web platform for managing all electrical equipment in factories, buildings, and offices. Experimental results show that our AIoT system achieves accurate recognition results.

关键词： Visualization Reactive power Buildings Machine learning Voltage Smart systems Real-time systems Production facilities Sensors Plugs

来源：评论

学校读者我要写书评

暂无评论

Non-Hermitian chiral degeneracy of gated graphene metasurfaces

引用

Light(Science & Applications) 2023年第5期12卷 766-775页

作者： Soojeong Baek Sang Hyun Park Donghak Oh Kanghee Lee Sangha Lee Hosub Lim Taewoo Ha Hyun Sung Park Shuang Zhang Lan Yang Bumki Min Teun-Teun Kim Department of Mechanical Engineering Korea Advanced Institute of Science and Technology(KAIST)Daehak-ro 291Daejeon 34141Republic of Korea Department of Electrical and Computer Engineering University of Minnesota200 Union street SEMinneapolisMN 55455USA Harvard Institute of Medicine Harvard Medical SchoolHarvard UniversityBrigham and Women's Hospital25 Shattuck StreetBostonMA 02215USA Center for Integrated Nanostructure Physics Institute for Basic Science(IBS)Sungkyunkwan UniversitySeobu-ro 2066Suwon 16419Republic of Korea Samsung Advanced Institute of Technology Samsung ElectronicsSamsung-roSuwon 16678Republic of Korea Department of Physics The University of Hong KongPokfulam RoadHong Kong Special Administrative RegionChina Department of Electrical and Systems Engineering Washington University1 Brookings DriveSaint LouisMO 63130USA Department of Physics Korea Advanced Institute of Science and Technology(KAIST)Daehak-ro 291Daejeon 34141Republic of Korea Department of Physics University of UlsanDaehakroUlsan 44610Republic of Korea Korea Research Institute of Standards and Science(KRISS) Gajeong-ro 267Daejeon 34113Republic of Korea

Non-Hermitian degeneracies,also known as exceptional points(EPs),have been the focus of much attention due to their singular eigenvalue surface ***,as pertaining to a non-Hermitian metasurface platform,the reduction of an eigenspace dimensionality at the EP has been investigated mostly in a passive repetitive ***,we propose an electrical and spectral way of resolving chiral EPs and clarifying the consequences of chiral mode collapsing of a non-Hermitian gated graphene *** specifically,the measured non-Hermitian Jones matrix in parameter space enables the quantification of nonorthogonality of polarisation eigenstates and half-integer topological charges associated with a chiral ***,the output polarisation state can be made orthogonal to the coalesced polarisation eigenstate of the metasurface,revealing the missing dimension at the chiral *** addition,the maximal nonorthogonality at the chiral EP leads to a blocking of one of the cross-polarised transmission pathways and,consequently,the observation of enhanced asymmetric polarisation *** anticipate that electrically controllable non-Hermitian metasurface platforms can serve as an interesting framework for the investigation of rich non-Hermitian polarisation dynamics around chiral EPs.

关键词： structure chiral integer

来源：评论

学校读者我要写书评

暂无评论

Control of Néel Vector with Spin-Orbit Torques in an Antiferromagnetic Insulator with Tilted Easy Plane

引用

Physical Review Letters 2022年第1期129卷 017203-017203页

作者： Pengxiang Zhang Chung-Tao Chou Hwanhui Yun Brooke C. McGoldrick Justin T. Hou K. Andre Mkhoyan Luqiao Liu Department of Electrical Engineering and Computer Science Massachusetts Institute of Technology Cambridge Massachusetts 02139 USA Department of Physics Massachusetts Institute of Technology Cambridge Massachusetts 02139 USA Department of Chemical Engineering and Materials Science University of Minnesota Minneapolis Minnesota 55455 USA

Injecting spin currents into antiferromagnets and realizing efficient spin-orbit-torque switching represents a challenging topic. Because of the diminishing magnetic susceptibility, current-induced antiferromagnetic dynamics remain poorly characterized, complicated by spurious effects. Here, by growing a thin film antiferromagnet, α−Fe2O3, along its nonbasal plane orientation, we realize a configuration where the spin-orbit torque from an injected spin current can unambiguously rotate and switch the Néel vector within the tilted easy plane, with an efficiency comparable to that of classical ferrimagnetic insulators. Our study introduces a new platform for quantitatively characterizing switching and oscillation dynamics in antiferromagnets.

关键词： Antiferromagnetism Spin dynamics Spin-orbit torque

来源：评论

学校读者我要写书评

暂无评论

Online MTPA Tracking for Optimal Performance of IPMSM based Compressor Drives 11

Online MTPA Tracking for Optimal Performance of IPMSM based ...

引用

11th International Conference on Power Electronics, Machines and Drives, PEMD 2022

作者： Chandrasekaran, Visweshwar Jose, Bernard Mohan, Ned Basu, Kaushik Trane Technologies St. PaulMN United States Trane Technologies Bangalore India Department of Electrical and Computer Engineering University of Minnesota MinneapolisMN United States Division of EECS Indian Institute of Sciences Bangalore India

ISBN: (数字)9781839537189

ISBN: (纸本)9781839537189

The efficiency of interior permanent magnet synchronous motors (IPMSM) can be directly influenced by Variable Frequency Drives (VFDs) implementing Maximum Torque Per Ampere (MTPA) algorithms. Model based optimization requires advanced knowledge of machine including non-linear properties which are quite pronounced in IPMSMs. This paper presents a novel online MTPA trajectory seeking scheme which is based on calculating the minimum current possible for every load operating point by dynamically adjusting the current angle during steady state operation. First, an offline MTPA trajectory is derived for the motor using simple datasheet parameters to get a baseline. Second, an online MTPA based on discrete extremum seeking state logic implemented using a model-based design approach is used to seek the least current for a given operating point. Key considerations have been made for efficient execution while minimizing computational resource requirements Experimental results obtained on a 3 HP IPMSM motor demonstrate the effectiveness of the proposed scheme in achieving the MTPA control objective while maintaining control stability across all operating regions. © IET Conference Proceedings. All right reserved.

关键词： Permanent magnets

来源：评论

学校读者我要写书评

暂无评论

Krylov Cubic Regularized Newton: A Subspace Second-Order Method with Dimension-Free Convergence Rate

arXiv

引用

arXiv 2024年

作者： Jiang, Ruichen Raman, Parameswaran Sabach, Shoham Mokhtari, Aryan Hong, Mingyi Cevher, Volkan Department of Electrical and Computer Engineering The University of Texas Austin United States Amazon Web Services United States Faculty of Data and Decision Sciences Technion Israel Institute of Technology Department of Electrical and Computer Engineering University of Minnesota United States LIONS IEM STI Ecole Polytechnique Fédérale de Lausanne Switzerland

Second-order optimization methods, such as cubic regularized Newton methods, are known for their rapid convergence rates;nevertheless, they become impractical in high-dimensional problems due to their substantial memory requirements and computational costs. One promising approach is to execute second-order updates within a lower-dimensional subspace, giving rise to subspace second-order methods. However, the majority of existing subspace second-order methods randomly select subspaces, consequently resulting in slower convergence rates depending on the problem’s dimension d. In this paper, we introduce a novel subspace cubic regularized Newton method that achieves a dimension-independent global convergence rate of O ( mk1 + k12 ) for solving convex optimization problems. Here, m represents the subspace dimension, which can be significantly smaller than d. Instead of adopting a random subspace, our primary innovation involves performing the cubic regularized Newton update within the Krylov subspace associated with the Hessian and the gradient of the objective function. This result marks the first instance of a dimension-independent convergence rate for a subspace second-order method. Furthermore, when specific spectral conditions of the Hessian are met, our method recovers the convergence rate of a full-dimensional cubic regularized Newton method. Numerical experiments show our method converges faster than existing random subspace methods, especially for high-dimensional problems. Copyright © 2024, The Authors. All rights reserved.

关键词： Newton-Raphson method

来源：评论

学校读者我要写书评

暂无评论

Challenges in Video-Based Infant Action Recognition: A Critical Examination of the State of the Art

arXiv

引用

arXiv 2023年

作者： Hatamimajoumerd, Elaheh Kakhaki, Pooria Daneshvar Huang, Xiaofei Luan, Lingfei Amraee, Somaieh Ostadabbas, Sarah Department of Electrical & Computer Engineering Northeastern University MA United States Roux Institute Northeastern University ME United States University of Minnesota MN United States

Automated human action recognition, a burgeoning field within computer vision, boasts diverse applications spanning surveillance, security, human-computer interaction, tele-health, and sports analysis. Precise action recognition in infants serves a multitude of pivotal purposes, encompassing safety monitoring, developmental milestone tracking, early intervention for developmental delays, fostering parent-infant bonds, advancing computer-aided diagnostics, and contributing to the scientific comprehension of child development. This paper delves into the intricacies of infant action recognition, a domain that has remained relatively uncharted despite the accomplishments in adult action recognition. In this study, we introduce a groundbreaking dataset called "InfActPrimitive", encompassing five significant infant milestone action categories, and we incorporate specialized preprocessing for infant data. We conducted an extensive comparative analysis employing cutting-edge skeleton-based action recognition models using this dataset. Our findings reveal that, although the PoseC3D model achieves the highest accuracy at approximately 71%, the remaining models struggle to accurately capture the dynamics of infant actions. This highlights a substantial knowledge gap between infant and adult action recognition domains and the urgent need for data-efficient pipeline models† © 2023, CC BY-NC-ND.

关键词： Human computer interaction

来源：评论

学校读者我要写书评

暂无评论

MADA: meta-adaptive optimizers through hyper-gradient descent 24

MADA: meta-adaptive optimizers through hyper-gradient descen...

引用

Proceedings of the 41st International Conference on Machine Learning

作者： Kaan Ozkara Can Karakus Parameswaran Raman Mingyi Hong Shoham Sabach Branislav Kveton Volkan Cevher Department of Electrical and Computer Engineering University of California Los Angeles Amazon Web Services Amazon Web Services and Department of Electrical and Computer Engineering University of Minnesota Faculty of Data and Decision Sciences Technion - Israel Institute of Technology Israel Amazon Web Services and LIONS IEM STI Ecole Polytechnique Fédérale de Lausanne Switzerland

Following the introduction of Adam, several novel adaptive optimizers for deep learning have been proposed. These optimizers typically excel in some tasks but may not outperform Adam uniformly across all tasks. In this work, we introduce Meta-Adaptive Optimizers (MADA), a unified optimizer framework that can generalize several known optimizers and dynamically learn the most suitable one during training. The key idea in MADA is to parameterize the space of optimizers and dynamically search through it using hyper-gradient descent during training. We empirically compare MADA to other popular optimizers on vision and language tasks, and find that MADA consistently outperforms Adam and other popular optimizers, and is robust against suboptimally tuned hyper-parameters. MADA achieves a greater validation performance improvement over Adam compared to other popular optimizers during GPT- 2 training and fine-tuning. We also propose AVGrad, a modification of AMSGrad that replaces the maximum operator with averaging, which is more suitable for hyper-gradient optimization. Finally, we provide a convergence analysis to show that parameterized interpolations of optimizers can improve their error bounds (up to constants), hinting at an advantage for meta-optimizers.

关键词：

来源：评论

学校读者我要写书评

暂无评论

MADA: Meta-Adaptive Optimizers through hyper-gradient Descent

arXiv

引用

arXiv 2024年

作者： Ozkara, Kaan Karakus, Can Raman, Parameswaran Hong, Mingyi Sabach, Shoham Kveton, Branislav Cevher, Volkan Department of Electrical and Computer Engineering University of California Los Angeles United States Amazon Web Services United States Department of Electrical and Computer Engineering University of Minnesota United States Faculty of Data and Decision Sciences Technion Israel Institute of Technology LIONS IEM STI Ecole Polytechnique Fédérale de Lausanne Switzerland

Following the introduction of Adam, several novel adaptive optimizers for deep learning have been proposed. These optimizers typically excel in some tasks but may not outperform Adam uniformly across all tasks. In this work, we introduce Meta-Adaptive Optimizers (MADA), a unified optimizer framework that can generalize several known optimizers and dynamically learn the most suitable one during training. The key idea in MADA is to parameterize the space of optimizers and dynamically search through it using hyper-gradient descent during training. We empirically compare MADA to other popular optimizers on vision and language tasks, and find that MADA consistently outperforms Adam and other popular optimizers, and is robust against sub-optimally tuned hyper-parameters. MADA achieves a greater validation performance improvement over Adam compared to other popular optimizers during GPT-2 training and fine-tuning. We also propose AVGrad, a modification of AMSGrad that replaces the maximum operator with averaging, which is more suitable for hyper-gradient optimization. Finally, we provide a convergence analysis to show that parameterized interpolations of optimizers can improve their error bounds (up to constants), hinting at an advantage for meta-optimizers. © 2024, CC BY.

关键词： Gradient methods

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：