检索结果-南通市图书馆

CSEE Journal of Power and Energy Systems 2024年第4期10卷 1563-1574页

作者： Chenxi Fan kaishun Xiahou Lei Wang Q.H.wu School of Electric Power Engineering South China University of TechnologyGuangzhou 510641China

This paper presents a long short-term memory(LSTM)-based fault detection method to detect the multiple open-circuit switch faults of modular multilevel converter(MMC)systems with full-bridge sub-modules(FB-SMs).Eighteen sensor signals of grid voltages,grid currents and capacitance voltages of MMC for single and multi-switch faults are collected as sampling *** output signal characteristics of four types of single switch faults of FB-SM,as well as double switch faults in the same and different phases of MMC,are analyzed under the conditions of load variations and control command changes.A multi-layer LSTM network is devised to deeply extract the fault characteristics of MMC under different faults and operation conditions,and a Softmax layer detects the fault *** results have confirmed that the proposed LSTM-based method has better detection performance compared with three other methods:K-nearest neighbor(KNN),naive bayes(NB)and recurrent neural network(RNN).In addition,it is highly robust to model uncertainties and Gaussian *** validity of the proposed method is further demonstrated by experiment studies conducted on a hardware-in-the-loop(HIL)testing platform.

关键词： Fault detection long short-term memory(LSTM) modular multilevel converter(MMC) open circuit fault

Rethinking the architecture design of data center networks

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Frontiers of Computer Science 2012年第5期6卷 596-603页

作者： kaishun wu Jiang XIAO Lionel M. NI National Engineering Research Center of Digital Life State-Province Joint Laboratory of Digital Home Interactive Applications School of Physics and Engineering Sun Yat-sen University Guangzhou 510006 China Department of Computer Science and Engineering Hong Kong University of Science and Technology Hong Kong China

In the rising tide of the Internet of things, more and more things in the world are connected to the Internet. Recently, data have kept growing at a rate more than four times of that expected in Moore＇s law. This explosion of data comes from various sources such as mobile phones, video cameras and sensor networks, which often present multidi- mensional characteristics. The huge amount of data brings many challenges on the management, transportation, and pro- cessing IT infrastructures. To address these challenges, the state-of-art large scale data center networks have begun to provide cloud services that are increasingly prevalent. How- ever, how to build a good data center remains an open chal- lenge. Concurrently, the architecture design, which signifi- cantly affects the total performance, is of great research inter- est. This paper surveys advances in data center network de- sign. In this paper we first introduce the upcoming trends in the data center industry. Then we review some popular design principles for today＇s data center network architectures. In the third part, we present some up-to-date data center frame- works and make a comprehensive comparison of them. Dur- ing the comparison, we observe that there is no so-called op- timal data center and the design should be different referring to the data placement, replication, processing, and query pro- cessing. After that, several existing challenges and limitations are discussed. According to these observations, we point out some possible future research directions.

关键词： data center networks switch-based networks,direct networks hybrid networks

Generative Neural Network Based Spectrum Sharing Using Linear Sum Assignment Problems

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

China Communications 2020年第2期17卷 14-29页

作者： Ahmed BZaky Joshua Zhexue Huang kaishun wu Basem MElHalawany Big Data Institute Shenzhen University518060China Benha University Benha13511Egypt Guangdong Laboratory of Artificial Intelligence and Digital Economy(SZ) Shenzhen UniversityShenzhen 518060China PCL Research Center of Networks and Communications Peng Cheng LaboratoryShenzhen 518060China

Spectrum management and resource allocation(RA)problems are challenging and critical in a vast number of research areas such as wireless communications and computer *** traditional approaches for solving such problems usually consume time and memory,especially for large-size *** different machine learning approaches have been considered as potential promising techniques for combinatorial optimization problems,especially the generative model of the deep neural *** this work,we propose a resource allocation deep autoencoder network,as one of the promising generative models,for enabling spectrum sharing in underlay device-to-device(D2D)communication by solving linear sum assignment problems(LSAPs).Specifically,we investigate the performance of three different architectures for the conditional variational autoencoders(CVAE).The three proposed architecture are the convolutional neural network(CVAECNN)autoencoder,the feed-forward neural network(CVAE-FNN)autoencoder,and the hybrid(H-CVAE)*** simulation results show that the proposed approach could be used as a replacement of the conventional RA techniques,such as the Hungarian algorithm,due to its ability to find solutions of LASPs of different sizes with high accuracy and very fast execution ***,the simulation results reveal that the accuracy of the proposed hybrid autoencoder architecture outperforms the other proposed architectures and the state-of-the-art DNN techniques.

关键词： autoencoder linear sum assignment problems generative models resource allocation

在线全文

学校读者我要写书评

暂无评论

面向云网融合的细粒度多接入边缘计算架构

计算机研究与发展 2021年第6期58卷 1275-1290页

作者：王璐张健浩王廷伍楷舜深圳大学计算机与软件学院广东深圳518060 上海市高可信计算重点实验室(华东师范大学) 上海200062

随着智能终端设备的爆发式增长,多接入边缘计算(multi-access edge computing,MEC)成为支持多服务、多租户生态系统的关键技术之一.多接入边缘计算通过结合云端的移动计算技术和接入网的无线通信技术,实现了云端和网络的高效融合.然而,... 详细信息

随着智能终端设备的爆发式增长,多接入边缘计算(multi-access edge computing,MEC)成为支持多服务、多租户生态系统的关键技术之一.多接入边缘计算通过结合云端的移动计算技术和接入网的无线通信技术,实现了云端和网络的高效融合.然而,目前的边缘计算技术对于所有可能的资源(例如计算、通信、缓存)并没有细粒度的控制能力,因此并不能对延迟敏感的实时服务提供很好的支持.为了解决这个关键问题,设计了一种基于软件定义(software defined)的细粒度多接入边缘计算架构,可以对网络资源和计算资源进行细粒度的控制并进行协同管理,并设计了一种基于深度强化学习Q-Learning的两级资源分配策略,从而提供更有效的计算卸载和服务增强.大量的仿真实验证明了该架构的有效性.

关键词：云网融合多接入边缘计算细粒度接入网软件定义网络深度强化学习

在线全文

学校读者我要写书评

暂无评论

一种基于GUI模型的测试脚本组合方法

东南大学学报（自然科学版） 2017年第A1期47卷 164-169页

作者：王进华黄松惠战伟吴开舜陆军工程大学南京210007 全军军事训练软件测评中心南京210007

为了提高测试效率和测试脚本复用率,提出了一种基于图形用户界面(graph user interface,GUI)模型的测试脚本组合方法.该方法针对GUI层次结构的特点,采用事件关系图和窗口调用图对GUI交互行为进行建模.首先,通过反射机制从程序可执行文... 详细信息

为了提高测试效率和测试脚本复用率,提出了一种基于图形用户界面(graph user interface,GUI)模型的测试脚本组合方法.该方法针对GUI层次结构的特点,采用事件关系图和窗口调用图对GUI交互行为进行建模.首先,通过反射机制从程序可执行文件中解析提取事件关系图和窗口调用图;然后,通过构建脚本对象分层信息,从而建立测试脚本与模型的映射关系,再利用映射关系与模型实现脚本的组合.最后,在4款开源GUI应用上进行实验.实验结果表明,所提方法在组合耗时、组合脚本可执行率和模型覆盖率3个方面具有明显优势,验证了脚本组合方法的有效性.

关键词：图形用户界面事件关系图窗口调用图测试脚本组合脚本映射

OESC：一种用于软件测试过程分析的分层框架

在线全文

学校读者我要写书评

暂无评论

东南大学学报（自然科学版） 2017年第A1期47卷 158-163页

作者：吴开舜黄松惠战伟王进华姚奕解放军理工大学指挥信息系统学院软件测评中心南京210007 军军事训练软件测评中心南京210007

针对使用测试管理工具难以准确地分析软件测试的执行过程这一问题,提出一种基于测试人员操作记录的分层框架(OESC框架).依据测试人员的行为粒度将该框架划分为操作层、事件层、语义层、概述层,其中操作层和事件层已有工具进行记录,语义... 详细信息

针对使用测试管理工具难以准确地分析软件测试的执行过程这一问题,提出一种基于测试人员操作记录的分层框架(OESC框架).依据测试人员的行为粒度将该框架划分为操作层、事件层、语义层、概述层,其中操作层和事件层已有工具进行记录,语义层和概述层是基于操作层和事件层内容,通过特征提取技术和多标签分类技术得到.实验结果表明,该框架对被测件性能影响极小,能有效地监控测试人员行为过程,有助于帮助分析测试过程,能弥补现有的测试管理工具对测试人员监控上的不足.

关键词：测试过程分析过程控制多标签分类用户行为

在线全文

学校读者我要写书评

暂无评论

基于深度学习的超高频标签识别系统

计算机科学 2023年第S1期50卷 656-661页

作者：余加宝姚俊梅谢瑞桃伍楷舜马军超深圳大学计算机与软件学院广东深圳518000 深圳技术大学大数据与互联网学院广东深圳518000

无线射频识别(Radio Frequency Identification,RFID)系统最基本的功能是标签识别,然而身份验证系统无法检测到伪造或克隆标签,从而出现潜在安全隐患和个人隐私问题。目前有基于加密的认证协议和基于特征提取的解决方法,其中基于加密的... 详细信息

无线射频识别(Radio Frequency Identification,RFID)系统最基本的功能是标签识别,然而身份验证系统无法检测到伪造或克隆标签,从而出现潜在安全隐患和个人隐私问题。目前有基于加密的认证协议和基于特征提取的解决方法,其中基于加密的认证协议方法不兼容现有的协议,基于特征提取的方法存在特征提取困难或者识别距离短等限制。文中基于标签物理层信号的真实性进行识别,结合深度学习技术,提出标签信号识别方法。其核心思想在于在RFID通信过程中,利用标签的后向散射信号提取与标签逻辑信息无关的信号,将提取的信号送入卷积神经网络进行相似度匹配,根据得到的相似度匹配分数与给定的阈值对比,最后实现标签的真实性识别。采用USRP N210作为RFID系统的阅读器,采用150个超高频商用标签作为信号的发射器,并在实际场景中采集真实的RFID信号。通过实验验证了基于深度学习的标签识别能达到94%以上的识别精度,在识别距离长达2 m的情况下其等错误比率(EER)为0.034。

关键词：物理层识别无线射频识别深度学习标签

在线全文

学校读者我要写书评

暂无评论

基于BM25算法的问题报告质量检测方法

清华大学学报（自然科学版） 2020年第10期60卷 829-836页

作者：陈乐乐黄松孙金磊惠战伟吴开舜中国人民解放军陆军工程大学指挥控制工程学院南京210007 全军军事软件测评中心南京210007

问题报告作为记录和跟踪缺陷的载体,为解决软件质量问题提供依据。目前软件测试常以多人、并行的方式进行,海量问题报告的去假与去重等整合过程正面临严峻的挑战。因此,该文提出一种基于BM25算法的问题报告自动化检测方法,在对问题报告... 详细信息

问题报告作为记录和跟踪缺陷的载体,为解决软件质量问题提供依据。目前软件测试常以多人、并行的方式进行,海量问题报告的去假与去重等整合过程正面临严峻的挑战。因此,该文提出一种基于BM25算法的问题报告自动化检测方法,在对问题报告进行预处理后,依据测试需求和测试报告样本建立匹配库,利用BM25算法计算两者的相似度得分,并以此为依据检测问题报告的正确性。在软件测试大赛的数据上进行实验,结果表明该文提出的方法能够正确评判大部分问题报告,有效提高了去假与去重效率。

关键词：软件测试 BM25算法问题报告自然语言处理

Constructing an Energy Function for Power Systems with DFIGWT Generation Based on a Synchronous-generator-mimicking Model

在线全文

学校读者我要写书评

暂无评论

CSEE Journal of Power and Energy Systems 2022年第1期8卷 64-75页

作者： Yang Liu Chenying Xu Zehui Lin kaishun Xiahou Q.H.wu the School of Electric Power Engineering South China University of TechnologyGuangzhou 510640China

In this paper we construct an energy function for multi-machine power systems with doubly-fed induction generator-based wind turbine(DFIGWT)according to a synchronous-generator-mimicking(SGM)model of the *** SGM model is proposed to approximate the dynamics of a *** to the modelling of a synchronous generator(SG),the internal dynamics of a DFIGWT are also described with differential equations of newly constructed virtual rotor angle and internal electromotive force(EMF)in the SGM ***,the power flow of a DFIGWT is expressed by nonlinear functions of its virtual rotor angle and internal *** SGM model bridges the gap between the irregular and complex modelling of DFIGWTs and the well-developed energy function construction techniques for SG *** on the SGM model,a numerical energy function is constructed for power systems with DFIGWT *** theoretical analysis and numerical studies were undertaken to validate that the proposed energy function satisfies the necessary conditions for an energy function of a power system.

关键词： DFIGWT energy function multi-machine power systems synchronous-generator-mimicking model

Detection of false data injection attacks on power systems using graph edge-conditioned convolutional networks

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Protection and Control of Modern Power Systems 2023年第2期8卷 1-12页

作者： Bairen Chen Q.H.wu Mengshi Li kaishun Xiahou School of Electric Power Engineering South China University of TechnologyGuangzhou 510641China

State estimation plays a vital role in the stable operation of modern power systems,but it is vulnerable to cyber *** data injection attacks(FDIA),one of the most common cyber attacks,can tamper with measure-ment data and bypass the bad data detection(BDD)mechanism,leading to incorrect results of power system state estimation(PSSE).This paper presents a detection framework of FDIA for PSSE based on graph edge-conditioned convolutional networks(GECCN),which use topology information,node features and edge *** deep graph architecture,the correlation of sample data is effectively mined to establish the mapping relationship between the estimated values of measurements and the actual states of power *** addition,the edge-conditioned convolution operation allows processing data sets with different graph *** studies are undertaken on the IEEE 14-bus system under different attack intensities and degrees to evaluate the performance of *** results show that GECCN has better detection performance than convolutional neural networks,deep neural net-works and support vector ***,the satisfactory detection performance obtained with the data sets of the IEEE 14-bus,30-bus and 118-bus systems verifies the effective scalability of GECCN.

关键词： Power system state estimation(PSSE) Bad data detection(BDD) False data injection attacks(FDIA) Graph edge-conditioned convolutional networks(GECCN)