检索结果-内蒙古大学图书馆

Validation of a PMU-based fault location identification method for smart distribution network with photovoltaics using real-time data

引用

IET GENERATION TRANSMISSION & DISTRIBUTION 2018年第21期12卷 5824-5833页

作者： Usman, Muhammad Usama Faruque, Md. Omar Florida State Univ Elect & Comp Sci 2000 Levy Ave Tallahassee FL 32310 USA Florida State Univ Ctr Adv Power Syst 2000 Levy Ave Tallahassee FL 32310 USA

A fault location (fl) identification method for smart distribution network is presented and validated using a digital real-time simulator (DRTS). The method can accurately identify the fl in a distribution network in the presence of distributed generation (DG). This method is based on state estimation (SE) algorithm which uses real-time data from simulated phasor measurement units (PMUs), placed in the distribution network. SE needs the fault currents of the generators and voltage measurements of an optimal number of nodes to perform the fl algorithm. The method was validated using the IEEE 37 node test feeder with DGs. PMUs are placed on the real-time model of the system. The real-time model was implemented on a DRTS which streams phasor data over the Internet using C37.118 protocol. OpenPDC is used to collect real-time PMU data coming from the DRTS. Microsoft SQL is used as a database management server to store data coming from OpenPDC. In the last step of the fl process, data stored in OpenPDC is fed into a fl identification algorithm to locate the fault. Both balanced and unbalanced fault types are applied to different nodes and an accurate estimation of the fl (over 90% of the cases) is achieved.

关键词： power engineering computing smart power grids phasor measurement fault location SQL distributed power generation protocols distribution networks power system measurement DG PMUs real-time model DRTS streams phasor data OpenPDC real-time PMU data real-time simulator store data Matlab-based fl identification algorithm balanced fault types unbalanced fault types PMU-based fault location identification method smart distribution network real-time data distributed generation state estimation algorithm SE simulated phasor measurement units fault currents generators voltage measurements fl algorithm IEEE 37 node test feeder operating distribution grid

来源：评论

学校读者我要写书评

暂无评论

Ensemble voting-based fault classification and location identification for a distribution system with microgrids using smart meter measurements

引用

IET SMART GRID 2023年第3期6卷 219-232页

作者： Islam, Md Maidul Usman, Muhammad Usama Newaz, Alvi Faruque, Md Omar FSU Elect & Comp Engn Tallahassee FL 32306 USA Idaho Natl Lab Idaho Falls ID USA

This study presents an ensemble learning approach for fault classification and location identification in a smart distribution network containing photovoltaics (PV)-based microgrid. Lack of available data points and the unbalanced nature of the distribution system make fault handling a challenging task for utilities. The proposed method uses event-driven voltage data from smart meters to classify and locate faults. The ensemble voting classifier is composed of three base learners;random forest, k-nearest neighbours, and artificial neural network. The fault location (fl) task has been formulated as a classification problem where the fault type is classified in the first step and based on the fault type, the faulty bus is identified. The method is tested on IEEE-123 bus system modified with added PV-based microgrid along with dynamic loading conditions and varying fault resistances from 0 to 20 omega for both unbalanced and balanced fault types. A further sensitivity analysis has been done to test the robustness of the proposed method under various noise levels and data loss errors in the smart meter measurements. The ensemble method shows improved performance and robustness compared to some previously proposed fl methods. Finally, the proposed method has been experimentally validated on a real-time simulation-based testbed using a state-of-the-art digital real-time simulator, industry standard DNP3 communication protocol and a cpu-based control centre running the fl algorithm.

关键词： base learners Distribution networks noise levels fault diagnosis Solar power stations and photovoltaic power systems Other learning models (inc. Naive Bayes) smart meters artificial neural network learning (artificial intelligence) ensemble voting-based fault classification varying fault resistances Power system protection real-time simulation-based testbed PV-based microgrid random forest smart distribution network Neural nets protocols IEEE-123 bus system sensitivity analysis nearest neighbour methods neural nets location identification smart meter measurements unbalanced fault types fault location distribution system ensemble learning approach distributed power generation digital real-time simulator data loss errors fault location task event-driven voltage data power distribution faults Power engineering computing ensemble voting classifier CPU-based control centre power engineering computing photovoltaic power systems ensemble method power generation faults industry standard DNP3 communication protocol dynamic loading conditions k-nearest neighbours Distributed power generation Power system measurement and metering random forests fl algorithm photovoltaics-based microgrid Combinatorial mathematics

来源：评论

学校读者我要写书评

暂无评论

Low-frequency dispersion in resistor-capacitor networks

引用

CHINESE JOURNAL OF PHYSICS 2021年 71卷 589-596页

作者： Benyahia, Ahmed Bouamrane, Rachid USTO MB Fac Phys LEPM BP 1505 El MNaouer Oran 31000 Algeria

In this paper, we modelled the dielectric response of binary disordered materials, which exhibit universal properties, using random RCC' networks. We used a bimodal distribution for the capacitive values unlike usual networks that are commonly found in the literature. The C value reflects the micro-capacitors formed by relatively large dielectric regions, while the C ' value reflects micro-capacitors formed by small dielectric regions. A low-power law of conductivity sigma ' was obtained;its exponent depends on the network's composition and the C '/C ratio. This was achieved using the well-known Frank and Lobb (fl) algorithm. We also provided an analytical formula of EMA for three-phase composites. Such an agreement of numerical results was obtained between the fl algorithm and EMA except at low and high frequencies, while a good agreement was obtained for low proportions. In addition, we attempted to model low-frequency dispersions (LFDs) using these RCC' networks. The simulated LFD results were typically good except for the real part of permittivity epsilon ' .

关键词： Composite materials Random RC networks Percolation Emergent response UDR LFD EMA fl algorithm

来源：评论

学校读者我要写书评

暂无评论

FedGrid: Federated Model Aggregation via Grid Shifting 16

FedGrid: Federated Model Aggregation via Grid Shifting

引用

16th IEEE/ACM International Conference on Utility and Cloud Computing (UCC) / 10th IEEE/ACM International Conference on Big Data Computing, Applications and Technologies (BDCAT)

作者： Kraychev, Boris Kiyamousavi, Ensiye Koychev, Ivan GATE Inst Sofia Bulgaria Sofia Univ St Kliment Ohridski Sofia Bulgaria

ISBN: (纸本)9798400702341

Federated Learning is a machine learning technique where independent devices (clients) cooperatively train a machine learning model by working on decentralized training data. A fundamental challenge in federated learning is the client model aggregation. The goal is to combine and preserve the knowledge that each client has acquired during its local training phase and generate an aggregated model with superior performance than any of the clients. Most current state-of-the-art algorithms for model aggregation still rely on simple weight parameter averaging. Though effective, these methods do not consider the internal structure of a deep neural network. In addition, the element-wise aggregation of the model parameters may create a conflict and loss of knowledge if a particular weight parameter serves different classification tasks over other learning clients. To decrease the probability of such conflicts, we propose FedGrid, a method that involves rotated (shifted) weight parameter grids during the clients' local training phase, which stimulates the models to build non-overlapping neural network structures. FedGrid can be applied to any existing algorithm that performs parameter averaging to improve the aggregation of deep NN models. To estimate the method's performance, we generate an ensemble of datasets with preset heterogeneity, measured via entropy balance metric, ranging from 0.1 to 1.0. Our results indicate that FedGrid outperforms the state-of-the-art methods when applied to deep neural network models. Moreover, the proposed method is simple and does not add computational complexity to the learning process. We believe that our effective method for grid application provides a new field for future research and improvement of the federation of deep neural network models.

关键词： Federated Learning Neural Networks Model Aggregation Strategy fl algorithm decentralized Learning Learning Rate Grid

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：