检索结果-内蒙古大学图书馆

Computational Optical Sensing and Imaging, COSI 2024 - Part of Optica Imaging Congress

作者： Markley, Eric Kabuli, Leyla Chien, Tiffany Pinkard, Henry Waller, Laura UCB/UCSF Joint Graduate Program in Bioengineering University of California Berkeley United States Department of Electrical Engineering and Computer Sciences University of California Berkeley United States

We present a technique for information-theoretic optimization of computational imaging systems demonstrated in snapshot 3D microscopy. By directly evaluating measurement quality and decoupling optimization from downst... 详细信息

关键词： Decoding

来源：评论

学校读者我要写书评

暂无评论

COMPARATIVE PERFORMANCE OF MACHINE LEARNING MODELS FOR TEMPERATURE PREDICTION IN VANADIUM REDOX FLOW BATTERIES

COMPARATIVE PERFORMANCE OF MACHINE LEARNING MODELS FOR TEMPE...

引用

International Conference on Green Energy, Computing and Intelligent Technology 2024, GEn-CITy 2024

作者： Zhang, Anxiu Juwono, Filbert H. Li, Hongfei Purwanto, Erick Wong, W.K. Liu, Linfeng Telecommunications Engineering Program Xi'an Jiaotong-Liverpool University Suzhou China Department of Electrical and Electronic Engineering Xi'an Jiaotong-Liverpool University Suzhou China Applied Mathematics Program Xi'an Jiaotong-Liverpool University Suzhou China Department of Computing Xi'an Jiaotong-Liverpool University Suzhou China Department of Electrical and Computer Engineering Curtin University Malaysia Miri Malaysia Sustainable Energy Technology Program Xi'an Jiaotong-Liverpool University Suzhou China

ISBN: (纸本)9781837242689

Vanadium Redox Flow Batteries (VRFB) are promising for large-scale energy storage due to their long life and environmental benefits. Accurate temperature prediction is key to optimizing VRFB performance and longevity. This study compares the performance of four machine learning models, i.e., 1D CNN, Particle Swarm Optimization - Support Vector Regressor (PSO-SVR), Decision Tree (DT), and K-Nearest Neighbors (KNN), using a publicly available dataset. Results show that KNN achieves the best results with test Root Mean Square Error (RMSE) of 0.0424 (average) and test R2 of 0.9746 (average), demonstrating strong predictive accuracy. 1D CNN, however, shows poor generalization. These findings suggest that non-parametric models like KNN and DT are highly effective for VRFB temperature prediction. © The Institution of engineering & Technology 2024.

关键词： machine learning prediction Vanadium redox flow battery

来源：评论

学校读者我要写书评

暂无评论

Smart Meter Application Analysis using PLS-SEM Deep Neural Network: A Case Study

引用

Journal of engineering Science and Technology Review 2023年第3期16卷 16-27页

作者： Thao, Nguyen Thi Phuong Duc, Minh Ly Bilik, Petr Faculty of Electrical and Electronics Engineering Ton Duc Thang University Ho Chi Minh City Viet Nam Faculty of Commerce Van Lang University Ho Chi Minh City Viet Nam Department of Cybernetics and Biomedical Engineering Faculty of Electrical Engineering and Computer Science VSB–Technical University of Ostrava 17. Listopadu2172/ 15 Ostrava708 00 Czech Republic

Smart meters measure, control, analyze, and predict the amount of electricity, water, and gas used. In developing countries, where there is no consensus to accept the use of smart meters, there are many possible risks when using smart meters. However, there are also many benefits of smart meters. This study conducts an overall assessment of the demand and impact of the smart meter in the southern region of Vietnam through a survey of 500 samples. This article examines information technology system (IS) related factors and engineering model-related factors according to technical readiness such as optimism, innovation, insecurity, and discomfort. Accompanying that is the expectation of smart meters, for Vietnamese people's intention to constantly use smart meters. Most of the previous studies on smart meter systems have focused on analyzing the impact of factors affecting the application using single-step structural equation modeling (SEM). In this study, it is proposed to use a 2-layer model between the research model of the multi-analysis method by combining the Partial Least Squares - Structural Equation Modeling (PLS-SEM) and Artificial Neural Networks (ANN) model was performed for additional analysis for the results of PLS-SEM, and ANN has higher predictive accuracy than PLS-SEM because ANN has the ability to perform well for both linear relational model and linear relationship model and non-linear with high prediction. First, the PLS-SEM model evaluates the factors affecting the intention to use the smart meter system. Second, the ANN ranks the impact factors of the critical predictors from the PLS-SEM model, and the Critical Performance Map Analysis (IPMA) analyzes the exact results for the critical performance of the variables elements. The results of this study show that the quality of information and quality of system factors of the IS model have a negative impact on users' intention to use smart meters. At the same time, factors of Optimization and innovati

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

High Gain 130-GHz Frequency Doubler with Colpitts Output Buffer Delivering Poutup to 8 dBm with 6% PAE in 55-nm SiGe BiCMOS

引用

IEEE Solid-State Circuits Letters 2021年 4卷 36-39页

作者： Pirbazari, Mahmoud M. Mazzanti, Andrea Department of Electrical Computer and Biomedical Engineering University of Pavia Pavia27100 Italy

A mm-Wave frequency doubler in an SiGe BiCMOS technology is presented. The core of the circuit comprises a push-push pair, for second-harmonic generation, and a stacked common-collector Colpitts oscillator which works as a common-base injection-locked amplifier to boost the conversion gain and output power. The class-C operation of the transistor in the Colpitts buffer leads to a pulsed current shape with enhanced second-harmonic content. As a result, the power conversion gain of the frequency doubler is increased by up to 10 dB, compared to the push-push pair alone. Moreover, the common-collector configuration keeps separate the oscillator tank from the load, allowing independent optimization of the harmonic conversion efficiency and the load impedance for maximum power delivery. Realized in an SiGe BiCMOS technology with 330-GHz f _{\mathrm{ max}} , the proposed frequency doubler delivers P _{\mathrm{ out}} up to 8 dBm at 130 GHz with 13-dB conversion gain and 6.3% power added efficiency. A Figure-of-Merit is proposed to benchmark frequency doublers and the presented chip shows up to 3 times improvement compared to the previously reported designs in the same frequency range. © 2018 IEEE.

关键词： Harmonic analysis

来源：评论

学校读者我要写书评

暂无评论

Improved Time-Localization of Power System Forced Oscillations Using Changepoint Detection

Improved Time-Localization of Power System Forced Oscillatio...

引用

2022 IEEE Power and Energy Society General Meeting, PESGM 2022

作者： Dosiek, Luke Hosur, Sanjay Union College Department of Electrical Computer and Biomedical Engineering SchenectadyNY12309 United States

ISBN: (数字)9781665408233

ISBN: (纸本)9781665408233

This paper explores the use of changepoint detection (CPD) for an improved time-localization of forced oscillations (FOs) in measured power system data. In order for the autoregressive moving average plus sinusoids (ARMA+S) class of electromechanical mode meters to successfully estimate modal frequency and damping from data that contains a FO, accurate estimates of where the FO exists in time series are needed. Compared to the existing correlation-based method, the proposed CPD method is based on upon a maximum likelihood estimator (MLE) for the detection of an unknown number changes in signal mean to unknown levels at unknown times. Using the pruned exact linear time (PELT) dynamic programming algorithm along with a novel refinement technique, the proposed approach is shown to provide a dramatic improvement in FO start/stop time estimation accuracy while being robust to intermittent FOs. These findings were supported though simulations with the minniWECC model. © 2022 IEEE.

关键词： Maximum likelihood estimation

来源：评论

学校读者我要写书评

暂无评论

Internet of Things based Smart Water Leakage Monitoring and Alert System 9

Internet of Things based Smart Water Leakage Monitoring and ...

引用

9th International Conference on Advanced Computing and Communication Systems, ICACCS 2023

作者： Keerthana, S. Elango, S. Judgi, T. Vivek, S.K. Ponmurugan, P. Kumar, Neelam Sanjeev J K K Nataraja College of Engineering and Technology Department of Electrical and Electronics Engineering Komarapalayam638183 India Nandha Engineering College Department of Electrical and Electronics Engineering Tamilnadu Erode India Sathyabama Institute of Science and Technology Department of Computer Science and Engineering India Shree Venkateshwara Hi-Tech Engineering College Department of Electrical and Electronics Engineering Gobichettipalayam638455 India Meenakshi Academy of Higher Education and Research Department of Research Chennai600078 India Saveetha School of Engineering SIMATS Department of Biomedical Engineering Chennai India

ISBN: (纸本)9798350397376

It is critical to manage water supply in a safe and efficient manner. Significant amounts of water are lost each year as a result of leaks in water distribution systems (WDN). As a result, locating leaks in a reliable manner is critical. Water leaks can be detected using monitoring software that analyses the flow of water. Apps that track the user's location can be extremely useful. If this technology is widely adopted, it will aid in the resolution of water management issues such as human error and waste of limited water resources. Water leaks can be located predictably using the Internet of Things (Internet of Things). Many countries struggle to get enough water, but the situation is not hopeless. It is critical to accurately locate water leaks in order to reduce system resistance. © 2023 IEEE.

关键词： Water distribution systems

来源：评论

学校读者我要写书评

暂无评论

Textual Training for the Hassle-Free Removal of Unwanted Visual Data: Case Studies on OOD and Hateful Image Detection 38

Textual Training for the Hassle-Free Removal of Unwanted Vis...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Lee, Saehyung Mok, Jisoo Park, Sangha Shin, Yongho Jung, Dahuin Yoon, Sungroh Department of Electrical and Computer Engineering Seoul National University Korea Republic of Qualcomm Korea YH Seoul Korea Republic of School of Computer Science and Engineering Soongsil University Korea Republic of Interdisciplinary Program in Artificial Intelligence Seoul National University Korea Republic of

In our study, we explore methods for detecting unwanted content lurking in visual datasets. We provide a theoretical analysis demonstrating that a model capable of successfully partitioning visual data can be obtained using only textual data. Based on the analysis, we propose Hassle-Free Textual Training (HFTT), a streamlined method capable of acquiring detectors for unwanted visual content, using only synthetic textual data in conjunction with pre-trained vision-language models. HFTT features an innovative objective function that significantly reduces the necessity for human involvement in data annotation. Furthermore, HFTT employs a clever textual data synthesis method, effectively emulating the integration of unknown visual data distribution into the training process at no extra cost. The unique characteristics of HFTT extend its utility beyond traditional out-of-distribution detection, making it applicable to tasks that address more abstract concepts. We complement our analyses with experiments in out-of-distribution detection and hateful image detection. Our codes are available at https://***/Saehyung-Lee/HFTT. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Dynamic Control of Reconfigurable Intelligent Surfaces: An IC-Based MOS Varactor Approach

Dynamic Control of Reconfigurable Intelligent Surfaces: An I...

引用

IEEE International Symposium on Circuits and Systems (ISCAS)

作者： Loukas Petrou Marco A. Antoniades Julius Georgiou Department of Electrical and Computer Engineering University of Cyprus Nicosia Cyprus Department of Electrical Computer and Biomedical Engineering Toronto Metropolitan University Ontario Canada

ISBN: (数字)9798350330991

ISBN: (纸本)9798350331004

Reconfigurable intelligent surfaces (RISs) are recognized as a fundamental enabler for improving energy efficiency in 6G and future networks. However, the power consumption and the reconfiguration delay still need improvement for what is required at GHz frequencies, thus delaying their commercial adaptation. On that regard, this study proposes the incorporation of Integrated Circuits (ICs) with MOS varactor loadings as part of the RIS framework, to improve power consumption and speed, while having precise tuning of the reflection phase for individual unit-cells. The presented circuit design features an asynchronous digital circuit responsible for transmitting binary streams to digital-to-analogue converters, which in turn, bias MOS varactors that are directly connected to each unit-cell within the RIS. The use of asynchronous digital control circuits facilitates the development of ultra-low power, high-speed ICs, thereby enhancing the dynamic scalability of the RIS system. Simulated results of the asynchronous circuit are presented on a mature, cost-effective, CMOS 0.18 μm process technology, showing static power consumption of 40,63 μW, dynamic energy consumption of 474.43 pJ and reconfiguration delay of 23.38 ns. The simulations are accompanied by a scalability analysis and a discussion of potential capabilities, offering valuable insights for the future of ICs on RIS systems. The proposed approach and circuit provide flexibility and performance to RIS systems not achievable with conventional control systems due to their benefits of using clockless networking communication.

关键词： Varactors Wireless communication Power demand Scalability Reconfigurable intelligent surfaces Energy efficiency Delays

来源：评论

学校读者我要写书评

暂无评论

Prospector Heads: Generalized Feature Attribution for Large Models & Data 41

Prospector Heads: Generalized Feature Attribution for Large ...

引用

41st International Conference on Machine Learning, ICML 2024

作者： Machiraju, Gautam Derry, Alexander Desai, Arjun Guha, Neel Karimi, Amir-Hossein Zou, James Altman, Russ B. Ré, Christopher Mallick, Parag Department of Biomedical Data Science Stanford University United States Cartesia AI United States Department of Computer Science Stanford University United States Department of Electrical & Computer Engineering University of Waterloo Canada Department of Radiology Stanford University United States

Feature attribution, the ability to localize regions of the input data that are relevant for classification, is an important capability for ML models in scientific and biomedical domains. Current methods for feature attribution, which rely on "explaining" the predictions of end-to-end classifiers, suffer from imprecise feature localization and are inadequate for use with small sample sizes and high-dimensional datasets due to computational challenges. We introduce prospector heads, an efficient and interpretable alternative to explanation-based attribution methods that can be applied to any encoder and any data modality. Prospector heads generalize across modalities through experiments on sequences (text), images (pathology), and graphs (protein structures), outperforming baseline attribution methods by up to 26.3 points in mean localization AUPRC. We also demonstrate how prospector heads enable improved interpretation and discovery of class-specific patterns in input data. Through their high performance, flexibility, and generalizability, prospectors provide a framework for improving trust and transparency for ML models in complex domains. Copyright 2024 by the author(s)

关键词： Data assimilation

来源：评论

学校读者我要写书评

暂无评论

Robotic Knee Tracking Control to Mimic the Intact Human Knee Profile Based on Actor-Critic Reinforcement Learning

引用

IEEE/CAA Journal of Automatica Sinica 2022年第1期9卷 19-30页

作者： Ruofan Wu Zhikai Yao Jennie Si He(Helen)Huang School of Electrical Computer and Energy EngineeringArizona State UniversityTempeAZ 85287 USA Department of Biomedical Engineering North Carolina State UniversityRaleighNC 27695 USA University of North Carolina at Chapel Hill Chapel HillNC 27599 USA

We address a state-of-the-art reinforcement learning(RL)control approach to automatically configure robotic pros-thesis impedance parameters to enable end-to-end,continuous locomotion intended for transfemoral amputee ***,our actor-critic based RL provides tracking control of a robotic knee prosthesis to mimic the intact knee *** is a significant advance from our previous RL based automatic tuning of prosthesis control parameters which have centered on regulation control with a designer prescribed robotic knee profile as the *** addition to presenting the tracking control algorithm based on direct heuristic dynamic programming(dHDP),we provide a control performance guarantee including the case of constrained *** show that our proposed tracking control possesses several important properties,such as weight convergence of the learning networks,Bellman(sub)optimality of the cost-to-go value function and control input,and practical stability of the human-robot *** further provide a systematic simulation of the proposed tracking control using a realistic human-robot system simulator,the OpenSim,to emulate how the dHDP enables level ground walking,walking on different terrains and at different *** results show that our proposed dHDP based tracking control is not only theoretically suitable,but also practically useful.

关键词： Automatic tracking of intact knee configuration of robotic knee prosthesis direct heuristic dynamic programming(dHDP) reinforcement learning control

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：