检索结果-内蒙古大学图书馆

Energy Technologies for Future Grids (ETFG), IEEE International Conference on

作者： Muamar Mohamed Farhad E. Mahmood Mehmmood A. Abd Miloud Rezkallah Abdelhamid Hamadi Ambrish Chandra Department of Electrical Engineering Ecole de Technologie Superieure University of Quebec Montreal QC Canada Department of Electrical Engineering University of Mosul Mosul Iraq Electrical and Computer Engineering IBM USA

This study developed a machine learning model to forecast electrical demand for smart micro-grids with the existing of numerical weather forecasting (NWP). The model uses three techniques, linear regression (LR), Long Short-Term Memory (LSTM) and eXtreme Gradient Boosting (XGBoost), to harness insights from historical data and predict future load patterns. Notably, the XGBoost algorithm is highlighted as the most effective due to underscoring the importance of feature selection and model optimization. The utilization of these techniques results in accurate electrical demand predictions, which are crucial for enhancing micro-grid efficiency, practically in response to weather-induced load fluctuations. This paper suggests employing advanced feature selection methods based on high feature importance, ultimately refining the model’s predictive capabilities for smart micro-grid applications.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Enforcing Privacy in Distributed Learning with Performance Guarantees

arXiv

引用

arXiv 2023年

作者： Rizk, Elsa Vlaski, Stefan Sayed, Ali H. The School of Engineering Ecole Polytechnique Fédérale de Lausanne Switzerland The Department of Electrical and Electronic Engineering Imperial College London United Kingdom

We study the privatization of distributed learning and optimization strategies. We focus on differential privacy schemes and study their effect on performance. We show that the popular additive random perturbation scheme degrades performance because it is not well-tuned to the graph structure. For this reason, we exploit two alternative graph-homomorphic constructions and show that they improve performance while guaranteeing privacy. Moreover, contrary to most earlier studies, the gradient of the risks is not assumed to be bounded (a condition that rarely holds in practice;e.g., quadratic risk). We avoid this condition and still devise a differentially private scheme with high probability. We examine optimization and learning scenarios and illustrate the theoretical findings through simulations. Copyright © 2023, The Authors. All rights reserved.

关键词： Privatization

来源：评论

学校读者我要写书评

暂无评论

High Data Throughput Wireless Networks Based on a Quantum Dash / Dot Multi-Wavelength Laser and Millimeter-Wave-Over-Fiber Transmission Links

High Data Throughput Wireless Networks Based on a Quantum Da...

引用

Photonics North

作者： Z.G. Lu G.C. Liu K. Zeb X.R. Xie J.R. Liu P.J. Poole Y.X. Mao P. Barrios M. Rahim M. Vachon C.Y. Song D. Poitras J. Weber N. Sabourin A. Askariann X.P. Zhang J.P. Yao K. Wu Advanced Electronics and Photonics Research Centre National Research Council Ottawa ON Canada Depart. Electrical and Computer Engineering iPhotonics Labs Concordia University Montreal QC Canada Department of Electrical Engineering Ecole Polytechnique Montreal Montreal QC Canada School of Electrical Engineering and Computer Science University of Ottawa Ottawa ON Canada

We have designed, epitaxy-grown, nano-fabricated and investigated several different-type InAs/InP quantum dash / dot (QD) multi-wavelength lasers (MWLs) around 1550 nm with very low relative intensity noise and ultra-narrow optical linewidth. By using those QD MWLs, we have experimentally demonstrated optical-heterodyne millimeter-wave-over-fiber fronthaul systems with high data throughput wireless links through 25- and 50-km single mode fiber featuring a free-space data capacity of up to 36 Gb/s, which has clearly indicated that the developed QD MWLs are critical building blocks of achieving low-noise millimeter-wave (mmW) signal generation and transmission for 5G & beyond wireless networks.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Effect of piezoelectric transducer material type on ultrasound assisted coalescence of water-in-oil emulsion 17

Effect of piezoelectric transducer material type on ultrasou...

引用

17th International Conference on Future Networks and Communications / 19th International Conference on Mobile Systems and Pervasive Computing / 12th International Conference on Sustainable Energy Information Technology, FNC/MobiSPC/SEIT 2022

作者： Adeyemi, Idowu Meribout, Mahmoud Khezzar, Lyes Kharoua, Nabil AlHammadi, Khalid Department of Mechanical Engineering Khalifa University of Science and Technology P.O. Box 127788 Abu Dhabi United Arab Emirates Department of Electrical Engineering and Computer Science Khalifa University of Science and Technology P.O. Box 127788 Abu Dhabi United Arab Emirates Department of Mechanical Engineering Ecole Nationale Polytechnique de Constantine Constantine Algeria

The usage of lead zirconate titanate poses serious health and environmental hazards. The production or utilization of PZTs results in the release of vaporized hazardous lead oxide (PbO) at elevated temperature or dissolution of poisonous lead into aqueous medium. Hence, in this study, the analysis of the effect of different piezoelectric material type (PZT, PVDF and Lithium Niobate) on domain acoustic pressure distribution and the subsequent coalescence of binary droplets was conducted. The lead-free piezoelectric materials (PVDF and Lithium Niobate) were assessed with the PZT in order to determine their viability for droplet coalescence. The piezoelectric materials were subjected to mechanical oscillation under resonance in order to allow for optimum performance. The eigen-frequency analysis provided the resonance frequencies of 19.61, 26.04 and 57.80 kHz for PVDF, PZT and Lithium niobate, respectively. Moreover, the acoustic pressure in the emulsion domain varied in magnitude and distribution. The order of the coalescence time for the binary droplets followed the order PZT © 2022 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (https://***/licenses/by-nc-nd/4.0) Peer-review under responsibility of the Conference Program Chairs.

关键词： Ultrasonics

来源：评论

学校读者我要写书评

暂无评论

Flexible Unsupervised Learning for Massive MIMO Subarray Hybrid Beamforming

Flexible Unsupervised Learning for Massive MIMO Subarray Hyb...

引用

GLOBECOM 2022 - 2022 IEEE Global Communications Conference

作者： Hamed Hojatian Jérémy Nadal Jean–François Frigon François Leduc–Primeau Department of Electrical Engineering Ecole Polytechnique de Montréal Montreal Quebec Canada

ISBN: (纸本)9781665435413

Hybrid beamforming is a promising technology to improve the energy efficiency of massive MIMO systems. In particular, subarray hybrid beamforming can further decrease power consumption by reducing the number of phase-shifters. However, designing the hybrid beamforming vectors is a complex task due to the discrete nature of the subarray connections and the phase-shift amounts. Finding the optimal connections between RF chains and antennas requires solving a non-convex problem in a large search space. In addition, conventional solutions assume that perfect channel state information (CSI) is available, which is not the case in practical systems. Therefore, we propose a novel unsupervised learning approach to design the hybrid beamforming for any subarray structure while supporting quantized phase-shifters and noisy CSI. One major feature of the proposed architecture is that no beamforming codebook is required, and the neural network is trained to take into account the phase-shifter quantization. Simulation results show that the proposed deep learning solutions can achieve higher sum-rates than existing methods.

关键词： Radio frequency Quantization (signal) Array signal processing Simulation Massive MIMO Energy efficiency Noise measurement

来源：评论

学校读者我要写书评

暂无评论

Fault Diagnosis in Robot Manipulators Using SVM and KNN

引用

Intelligent Automation & Soft Computing 2023年第2期35卷 1957-1969页

作者： D.Maincer Y.Benmahamed M.Mansour Mosleh Alharthi Sherif S.M.Ghonein Laboratoire de Robotiques Parallélisme et Systèmes Embarquer Département Automatique et ControlUniversitédes Sciences et Technologie Houari BoumedieneAlgiersAlgeria Laboratoire de Recherche en Electrotechnique Ecole Nationale PolytechniqueB.P 182El-Harrach16200AlgiersAlgeria Electrical Engineering Department College of EngineeringTaif UniversityP.O.Box 11099Taif21944Saudi Arabia

In this paper,Support Vector Machine(SVM)and K-Nearest Neighbor(KNN)based methods are to be applied on fault diagnosis in a robot manipulator.A comparative study between the two classiﬁers in terms of successfully detecting and isolating the seven classes of sensor faults is considered in this *** both classiﬁers,the torque,the position and the speed of the manipulator have been employed as the input ***,it is to mention that a large database is needed and used for the training and testing *** SVM method used in this paper is based on the Gaussian kernel with the parametersγand the penalty margin parameter“C”,which were adjusted via the PSO algorithm to achieve a maximum accuracy *** were carried out on the model of a Selective Compliance Assembly Robot Arm(SCARA)robot manipulator,and the results showed that the Particle Swarm Optimization(PSO)increased the per-formance of the SVM algorithm with the 96.95%accuracy while the KNN algo-rithm achieved a correlation up to 94.62%.These results showed that the SVM algorithm with PSO was more precise than the KNN algorithm when was used in fault diagnosis on a robot manipulator.

关键词： Support Vector Machine(SVM) Particle Swarm Optimization(PSO) K-Nearest Neighbor(KNN) fault diagnosis manipulator robot(SCARA)

来源：评论

学校读者我要写书评

暂无评论

Revisiting Character-level Adversarial Attacks for Language Models

arXiv

引用

arXiv 2024年

作者： Rocamora, Elias Abad Wu, Yongtao Liu, Fanghui Chrysos, Grigorios G. Cevher, Volkan LIONS École Polytechnique Fédérale de Lausanne Switzerland Department of Computer Science University of Warwick United Kingdom Department of Electrical and Computer Engineering University of Wisconsin-Madison United States

Adversarial attacks in Natural Language Processing apply perturbations in the character or token levels. Token-level attacks, gaining prominence for their use of gradient-based methods, are susceptible to altering sentence semantics, leading to invalid adversarial examples. While character-level attacks easily maintain semantics, they have received less attention as they cannot easily adopt popular gradient-based methods, and are thought to be easy to defend. Challenging these beliefs, we introduce Charmer, an efficient query-based adversarial attack capable of achieving high attack success rate (ASR) while generating highly similar adversarial examples. Our method successfully targets both small (BERT) and large (Llama 2) models. Specifically, on BERT with SST-2, Charmer improves the ASR in 4.84% points and the USE similarity in 8% points with respect to the previous art. Our implementation is available in ***/LIONS-EPFL/Charmer. © 2024, CC BY.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Adaptive Fractional-Order Sliding Mode Control for Mobile Robots: A PSO-Based Optimization Framework

Adaptive Fractional-Order Sliding Mode Control for Mobile Ro...

引用

IEEE SSD International Multi-Conference on Systems, Signals and Devices

作者： M. Nasir R. Fareh J. Ghommam Om Gad S. Khadraoui M. Bettayeb Department of Electrical Engineering University of Sharjah Sharjah UAE Department of Electrical and Computer Engineering Sultan Qaboos University Oman Control and Energy Management Laboratory (CEM Lab) Ecole National d'Ingenieur de Sfax Tunisia

ISBN: (数字)9798331542726

ISBN: (纸本)9798331542733

This study presents an adaptive fractional sliding mode control (AFSMC) strategy for finite-time trajectory tracking of differential drive mobile robots (DDMR) in the presence of system uncertainties and external disturbances. The proposed control strategy employs a fractional-order sliding surface and an optimized reaching law to ensure finite-time convergence of tracking errors while retaining robustness against external disturbances. Particle Swarm Optimization (PSO) is employed for offline parameter tuning, where its capability to efficiently handle a large parameter space makes it an ideal choice for optimizing the control system, ensuring both enhanced performance and computational feasibility. Simulation tests comparing the AFSMC with a conventional SMC show that the AFSMC exhibits higher tracking accuracy, robustness, and precise linear and angular velocity tracking. The results demonstrate the AFSMC's resilience to disturbances and minimal steady-state error, even under challenging conditions.

关键词： Adaptive systems Uncertainty Accuracy Torque Trajectory tracking Robustness Mobile robots Sliding mode control Optimization Tuning

来源：评论

学校读者我要写书评

暂无评论

Performance Evaluation of Multi-Modulation Single-Carrier PWM Methods for Motor Drive Applications

Performance Evaluation of Multi-Modulation Single-Carrier PW...

引用

International Conference on Power Electronics and Drive Systems (PEDS)

作者： P. M. Lingom J. Song-Manguelle Roland Unruh J. M Nyobe-Yome Mamadou Lamine Doumbia Department of Electrical and Computer Engineering University of Quebec at Trois-Rivieres Trois-Rivières Canada Power Electronics and Electrical Drives Paderborn University Paderborn Germany Ecole Nationale Polytechnique de Douala University of Douala Cameroon

Due to their numerous benefits, multi-modulation single-carrier pulse width modulation (MSC-PWM) approaches are becoming more prevalent in multilevel converter topology control systems. Compared to traditional multi-carrier PWM techniques, MSC-PWM methods present more straightforward real-time implementation, while achieving acceptable harmonic performances. However, their application remains limited to grid integration of photovoltaic converter systems, and no literature examines their suitability and effectiveness in multilevel converters supplying electrical machines. This paper uses the traditional and modified MSC-PWM methods to discuss the design and implementation of vector control for an induction motor fed by a multilevel cascaded H-bridge inverter. First, the vector control scheme is briefly introduced, and then each MSC-PWM method’s implementation scheme is thoroughly described. Then, the effectiveness of the investigated MSC-PWM methods is evaluated through steady-state and dynamic computer simulation results. Finally, a performance comparison of different PWM methods is conducted based on the spectra and THDs of the motor voltage, current, and electromagnetic torque. According to the comparison study, the modified MSC-PWM technique is superior to the conventional MSC-PWM method in reducing the motor voltage THD by up to 30%, motor current THD by up to 40%, and subsequently reducing the torque ripple in the motor air gap.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Optical sensing of charge and spin current fluctuations in centrosymmetric semiconductors

arXiv

引用

arXiv 2024年

作者： Lakhal, Amin Virally, Stéphane Khurgin, Jacob B. Seletskiy, Denis V. FemtoQ Laboratory Department of Engineering Physics Polytechnique Montréal MontréalQC Canada Electrical and Computer Engineering Department Johns Hopkins University BaltimoreMD21218 United States

We propose a time-resolved optical measurement scheme for sampling transient charge and spin currents in a bulk centrosymmetric semiconductor. The technique relies on emission of second harmonic light triggered by a pulsed below-gap optical excitation and a spontaneous intraband polarization arising from spin or charge motion, mediated by a χ(3)-based nonlinear optical process. Our proposal uses homodyne amplification to boost the weak second harmonic signal, making it detectable with conventional electronics, calculated for charge current in a room temperature GaAs semiconductor. This all-optical technique requires neither electrical contact nor bias fields and the signal is estimated at a few percent relative to the shot noise of the probe. This proposal motivates a novel method for exploring thermal and quantum fluctuations in the solid state in a non-invasive manner. © 2024, CC BY.

关键词： Shot noise

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：