Dissertation/Thesis Abstract

Date-Driven Topology Identification in Power Distribution Systems with Machine Learning
by Li, Yifu, M.S., The George Washington University, 2020, 129; 28087793
Abstract (Summary)

随着对优质电力需求的增加以及用户数量的增长,电网的运行和控制变得越来越复杂和充满挑战。确保基本的可靠性和供电质量在电气化经济的各个方面都变得尤为重要。随着微相量测量单元(μPMU)在配电网中的应用日益广泛,大量的高分辨率可检测量现可用于配电网的智能响应和故障分析。传统的一些模式暴露出电网拓扑识别的局限性,即在面临故障和其他干扰项时,往往会占用人力物力资源,却无法保证在短时间内有效恢复电力。本文提出并实现了一个以μPMU测量值作为输入的机器学习框架,能够实时地对配电网拓扑结构进行全面观察。具体地说,该框架采用卷积神经网络(CNN)来实时识别电网的参数状态。该框架在IEEE 34节点测试馈线上进行了实验验证,结果表明即使μPMU测量量包含噪声数据或数据丢失,该框架下的CNN仍能拥有较高精度的拓扑识别性能。

Indexing (document details)
Advisor: Dehghanian, Payman
Commitee: Ahmadi, Shahrokh, Imani, Mahdi
School: The George Washington University
Department: Electrical Engineering
School Location: United States -- District of Columbia
Source: MAI 82/2(E), Masters Abstracts International
Source Type: DISSERTATION
Subjects: Electrical engineering, Artificial intelligence, Information Technology
Keywords: Convolutional Neural Network, Micro-Phasor Measurement Unit (μPMU), Power distribution network, Topology identification
Publication Number: 28087793
ISBN: 9798664738728
Copyright © 2020 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy
ProQuest