检索结果-内蒙古大学图书馆

Emergency repair strategy of distribution network under wind disaster based on multidimensional importance evaluation

引用

ENERGY REPORTS 2023年 9卷 275-282页

作者： Wu, Wenjie Hou, Hui Wu, Xixiu Wei, Ruizeng He, Huan Shi, Ying Wuhan Univ Technol Sch Automat Wuhan 430070 Peoples R China Guangdong Power Grid Co Ltd Elect Power Res Inst Guangdong Key Lab Elect Power Equipment Reliabil Guangzhou 510080 Peoples R China

In order to enhance the power supply reliability of distribution network under wind disasters and provide assistance for distribution network operators to formulate post disaster repair strategies, this paper proposes an optimization method for key nodes and lines evaluation and distribution network repair strategy under wind disaster. Firstly, based on the complex network theory, we considered the status of nodes and lines in the topology and working state. The multi-dimensional evaluation indexes are proposed including improved node degree, node cohesion degree, node intermediate number, weighted line intermediate number, line transfer power flow entropy and line failure rate, etc. Then, the entropy weight method is used to objectively weight the six indicators, and TOPSIS algorithm is used to rank the comprehensive importance of nodes and lines in the distribution network. On this basis, the dynamic programming theory is used to optimize the emergency repair strategy of the fault distribution network after the wind disaster. And the problem of restoring as many key nodes and lines as possible in a short time is proposed. Finally, the method is applied to the ieee 39 bus system network to verify the feasibility. (c) 2023 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://***/licenses/by/4.0/).

关键词： Wind disaster Complex network theory TOPSIS Dynamic programming Rush to repair

来源：评论

学校读者我要写书评

暂无评论

NAR-Former: Neural Architecture Representation Learning towards Holistic Attributes Prediction

NAR-Former: Neural Architecture Representation Learning towa...

引用

ieee/CVF conference on Computer Vision and Pattern Recognition (CVPR)

作者： Yi, Yun Zhang, Haokui Hu, Wenze Wang, Nannan Wang, Xiaoyu Xidian Univ Xian Peoples R China Intellifusion Shenzhen Peoples R China Harbin Inst Technol Shenzhen Peoples R China

ISBN: (纸本)9798350301298

With the wide and deep adoption of deep learning models in real applications, there is an increasing need to model and learn the representations of the neural networks themselves. These models can be used to estimate attributes of different neural network architectures such as the accuracy and latency, without running the actual training or inference tasks. In this paper, we propose a neural architecture representation model that can be used to estimate these attributes holistically. Specifically, we first propose a simple and effective tokenizer to encode both the operation and topology information of a neural network into a single sequence. Then, we design a multi-stage fusion transformer to build a compact vector representation from the converted sequence. For efficient model training, we further propose an information flow consistency augmentation and correspondingly design an architecture consistency loss, which brings more benefits with less augmentation samples compared with previous random augmentation strategies. Experiment results on NAS-Bench101, NAS-Bench-201, DARTS search space and NNLQP show that our proposed framework can be used to predict the aforementioned latency and accuracy attributes of both cell architectures and whole deep neural networks, and achieves promising performance. Code is available at https://***/yuny220/NAR-Former.

关键词： Deep learning architectures and techniques

来源：评论

学校读者我要写书评

暂无评论

Poster Abstract: Text2Net: Transforming Plain Text into Dynamic, Interactive network Simulations 23

Poster Abstract: Text2Net: Transforming Plain Text into Dyna...

引用

23rd ACM/ieee International conference on Information Processing in Sensor networks, IPSN 2024

作者： Marefat, Alireza Mohamed Nishar, Abbaas Alif Ashok, Ashwin Georgia State Univerisity AtlantaGA United States

ISBN: (纸本)9798350362015

This paper introduces Text2Net, an innovative system designed to transform plain English descriptions into dynamic, interactive network simulations within the Emulator Virtual Engine-Next Generation (Eve-NG) environment. By integrating SOTA technologies from Natural Language Processing (NLP), Large Language Models (LLMs), and proposed adaptor software, Text2Net bridges the technical knowledge gap, enabling both technical and non-technical users to effortlessly create and interact with complex network topologies within a simulation environment. The system architecture combines an intuitive chat interface, GPT4 LLM to interpret user inputs, NLP key-value extraction, a simulation adaptor, and the EVE-NG engine. TextNet democratizes network emulation, empowering educators to efficiently construct simulations for interactive learning. It also benefits industrial prototyping and testing *** CONCEPTS•Applied computing → Interactive learning environments;Computer-assisted instruction;IT architectures;•networks → network design principles;programming interfaces;Topology analysis and generation;Logical / virtual topologies;network manageability;Programmable networks;network management;network monitoring. © 2024 ieee.

关键词： Engines

来源：评论

学校读者我要写书评

暂无评论

Stochastic Spiking Attention: Accelerating Attention with Stochastic Computing in Spiking networks 6

Stochastic Spiking Attention: Accelerating Attention with St...

引用

6th International conference on AI Circuits and Systems (AICAS)

作者： Song, Zihang Katti, Prabodh Simeone, Osvaldo Rajendran, Bipin Kings Coll London Dept Engn London WC2R 2LS England

ISBN: (纸本)9798350383638;9798350383645

Spiking Neural networks (SNNs) have been recently integrated into Transformer architectures due to their potential to reduce computational demands and to improve power efficiency. Yet, the implementation of the attention mechanism using spiking signals on general-purpose computing platforms remains inefficient. In this paper, we propose a novel framework leveraging stochastic computing (SC) to effectively execute the dot-product attention for SNN-based Transformers. We demonstrate that our approach can achieve high classification accuracy (83.53%) on CIFAR-10 within 10 time steps, which is comparable to the performance of a baseline artificial neural network implementation (83.66%). We estimate that the proposed SC approach can lead to over 6.3x reduction in computing energy and 1.7x reduction in memory access costs for a digital CMOS-based ASIC design. We experimentally validate our stochastic attention block design through an FPGA implementation, which is shown to achieve 48x lower latency as compared to a GPU implementation, while consuming 15x less power.

关键词： Spiking neural network Transformer attention stochastic computing hardware accelerator

来源：评论

学校读者我要写书评

暂无评论

Analysis and Design of a Compute-in-Memory System for Epileptic Seizure Detection System 67

Analysis and Design of a Compute-in-Memory System for Epilep...

引用

67th ieee International Midwest Symposium on Circuits and Systems (MWSCAS)

作者： Gulfaraz, Mobeen Johnson, Sam Saadeh, Wala Lahore Univ Management Sci LUMS Elect Engn Dept Lahore Pakistan Western Washington Univ WWU Engn & Design Dept Bellingham WA USA

ISBN: (纸本)9798350387186;9798350387179

This paper presents an area and power-efficient design of a dual-channel electroencephalogram (EEG) based digital processor that can identify the onset of seizures with high accuracy. Multiple modules, including feature extraction and classification, constitute the hardware design. Neural networks are employed as classifiers because of their exceptional performance. These networks are data-driven algorithms that cause the system's performance to deteriorate and its power consumption to increase. A compute-in-memory (CIM) architecture has thus been used to address this problem. The performance of the neural network algorithm is evaluated using an open-source database of CHBMIT with 87.1% accuracy, 91.2% sensitivity, and 90.5% specificity. In contrast, the performance of hardware architecture was evaluated using its power, area, and latency. The synthesis and implementation were carried out using the 0.18 mu m CMOS process. A comparison of two architectures, the fixed point neural network architecture and the CIM neural network architecture, clearly indicated that the CIM processor performed significantly better in terms of power and area while sacrificing latency because of bit serial computing. To detect seizures in real-time, the CIM processor consumes 241 mu J/classification while the fixed-point neural network consumes 256 mu J/classification of energy.

关键词： Epilepsy Electroencephalogram (EEG) Neural network Compute-in-memory (CIM)

来源：评论

学校读者我要写书评

暂无评论

Research on dynamic network security detection technology based on SDN 4

Research on dynamic network security detection technology ba...

引用

4th International conference on network Communication and Information Security, ICNCIS 2024

作者： Zhao, Kai Beijing Polytechnic Beijing100176 China

ISBN: (数字)9781510688261

ISBN: (纸本)9781510688254

Traditional network architecture can no longer meet the development needs of technologies such as cloud computing and big data. SDN network architecture has the characteristics of high openness and programmability, which can quickly respond to changes in business requirements. SDN decouples the control management function of the network from the data forwarding function, simplifies network configuration work based on logic centralized controllers and open programming interfaces, thereby achieving a flat management mode and flexible data forwarding function. The centralized control and programmability of SDN networks also pose significant security threats to SDN. Based on the CSE-CIC-IDS2018 dataset, this article tests the performance of various intelligent learning algorithms, analyzes data flows, distinguishes normal data from abnormal data, improves detection accuracy, and has practical significance for the security defense of SDN networks. © 2025 SPIE.

关键词： network security Computer security network architectures Control systems Machine learning Data transmission Defense and security Detection and tracking algorithms Information security Security technologies

来源：评论

学校读者我要写书评

暂无评论

Deep Neural network Representation for Explainable Machine Learning Algorithms: A Method for Hardware Acceleration

Deep Neural Network Representation for Explainable Machine L...

引用

ieee International Instrumentation and Measurement Technology conference (I2MTC)

作者： Schauer, Julian Goodarzi, Payman Schuetze, Andreas Schneider, Tizian Saarland Univ Lab Measurement Technol Saarbrucken Germany

ISBN: (纸本)9798350380903;9798350380910

Deep Learning and Machine Learning algorithms achieved good results in various tasks, from computer vision and speech recognition to condition monitoring and predictive maintenance. Besides the good results obtained, most Deep Learning algorithms do not provide a precise interpretation of their decisions. For the purpose of interpretability, explainable ML has gained attention. Because hardware accelerators do not support most explainable Machine Learning algorithms, they frequently experience runtime deficiencies. The resulting need for comparably inefficient software implementations limits their use on edge-computing hardware. This paper presents a method to represent trained explainable ML algorithms as Deep Neural networks to utilize hardware acceleration to meet the growing demand for accelerated inference of Machine Learning algorithms on edge hardware. The primary approach of the paper is to disassemble the trained explainable ML algorithms' inference into their basic mathematical operations to represent them as Deep Neural network layers. The techniques to convert the trained model functionalities to Deep Neural network layers are described in detail, including the layer functionalities and their usage in the Deep Neural networks. Due to the wide use of Deep Neural networks on hardware accelerators, this method allows the usage of affordable and efficient edge hardware instead of high-price, customized hardware and replaces the programming of high-effort compiler. Finally, the method is successfully applied to a part of an open-source ML toolbox, and the resulting Deep Neural network inferences are successfully run on a Neural Processing Unit.

关键词： edge computing hardware acceleration machine learning neural network

来源：评论

学校读者我要写书评

暂无评论

Dynamic Traffic Load Rebalancing for Hardware-accelerated 6G UPF Resilient Architecture

Dynamic Traffic Load Rebalancing for Hardware-accelerated 6G...

引用

2024 International Performance Computing and Communications conference

作者： Lee, DongJin Park, Seongsu Lee, BumJun Core R&D Dev ICT Infra Tech SK Telecom Gyeonggi South Korea Newgens R&D Dept Gyeonggi South Korea

ISBN: (纸本)9798350367959;9798350367942

Recently, research on new 6G mobile network architectures has started to explore how to provide AI-enabling Infra using hardware accelerators in a cloud-native environment. User Plane Function (UPF) in mobile Core network, which handles multi-Tbps packet data from millions of mobile subscribers, could benefit from such offloading. However, UPF must be highly robust and resilient to provide the best service quality to end-users even under high traffic surge situations. We present a dynamic traffic load rebalancing mechanism in UPF using Integer Linear programming (ILP) framework, to ensure even packet processing load allocation among Data Processing Units (DPUs). Our analysis and simulation results show the efficacy of the proposed scheme, demonstrating its potential for improved resiliency of hardware accelerated 6G UPF architecture.

关键词： 6G Core network UPF Architecture Rebalancing Optimization

来源：评论

学校读者我要写书评

暂无评论

FGNN-based Improved Resource Distribution Framework for V2X Wireless networks 99

FGNN-based Improved Resource Distribution Framework for V2X ...

引用

ieee 99th Vehicular Technology conference (VTC-Spring)

作者： Waqas, Syed Muhammad Tang, Yazhe Abbas, Fakhar Hussain, Mehboob Bangash, Yawar Abbas Xi An Jiao Tong Univ Sch Comp Sci & Technol Xian 710049 Peoples R China Natl Univ Singapore Nus Ctr Trusted Internet & Community Singapore Singapore Southwest Jiaotong Univ Sch Informat Sci & Technol Chengdu 610031 Peoples R China Natl Univ Sci & Technol NUST Dept Comp Software Engn Islamabad Pakistan

ISBN: (纸本)9798350387414

Recently, deep learning has emerged as a promising approach for solving challenging resource distribution (RD) problems in vehicle-to-everything (V2X) wireless networks. However, existing neural network architectures lack scalability, interpretability, and generalization. To address these limitations, in this study, we propose a new flexible graph neural network (FGNN)-based resource distribution framework for vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) user selection and power management in V2X networks with several next-generation access points (APs) and a cluster of V2V and V2I communication users. In the proposed framework, we formulated an optimization problem with each V2V and V2I user with the least power constraint that adapts to V2X wireless network settings through training inactive users. Likewise, we consider the situation when every V2I user shares the band with a different group of V2V users. Moreover, we introduce a parameterization of the RD framework strategy employing a flexible graph neural network (FGNN) context derived from instantaneous channel conditions to learn the low-dimension features of every user/vehicle. To assess the execution of the framework, we conduct simulation experiments comparing it with baseline methods in terms of efficiency, sum rate, and fairness.

关键词： Deep learning Resource distribution Graph neural network (GNN) V2X communication Wireless networks

来源：评论

学校读者我要写书评

暂无评论

Development of Intelligent Ground Control Station and Autopilot architectures for UAVs and UGVs 32

Development of Intelligent Ground Control Station and Autopi...

引用

32nd ieee Signal Processing and Communications Applications conference (SIU)

作者： Cintas, Emre Ozyer, Baris Ataturk Univ Bilgisayar Muhendisligi Bolumu Erzurum Turkiye Erzurum Tekn Univ Bilgisayar Muhendisligi Bolumu Erzurum Turkiye

ISBN: (纸本)9798350388978;9798350388961

Multi-rotor Unmanned Aerial and Ground Vehicles (UAV/UGV) are utilized for various military and civilian purposes such as detection, mapping, surveillance, target destruction, observation, and logistical applications due to their high agility, mechanical simplicity, ease of control, and autonomous capabilities. With advancing technology, UAVs and UGVs can be remotely controlled autonomously by onboard computers or by a pilot at a ground control station (GCS). Integrating these onboard computers into UAVs/UGVs enables information flow between the different modules they possess and the GCS. However, due to the traditional processing of this information flow in GCS software studies in the literature, there is a need for different and intelligent GCS software and autopilot design approaches. In this study, an intelligent new GCS capable of remotely commanding and controlling UAVs/UGVs has been designed, equipped with artificial intelligence features, capable of real-time recognition, positioning of more than seventy objects, and communication with UAVs/UGVs. The open-source Java programming language was used for coding operations, and the MAVLink (Micro Air Vehicle Link) protocol was utilized for communication between UAVs/UGVs and the GCS. For real-time object detection and localization, one of the fast and highly accurate algorithms in the literature, SSD MobileNet v2, was used and integrated with the GCS via Flask API. The obtained results demonstrated the consistency between the initial motivation and the software outputs, leading to the development of an open-source intelligent ground control station software and autopilot.

关键词： Unmanned aerial and ground vehicle Intelligent ground control station Computer vision Autopilot MAVLink

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：