Applications of machine learning techniques to the design of a self-healing smart grid
碩士 === 國立東華大學 === 電機工程學系 === 101 === Recently, renewable energy has been promoted all over the world, and the establishment of smart meters and smart grids are gradually becoming a new type of power grid. Although a smart grid gradually forms nowadays, the part of power interruption during transmi...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Published: |
2013
|
Online Access: | http://ndltd.ncl.edu.tw/handle/04870825052990840915 |
id |
ndltd-TW-101NDHU5442052 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-101NDHU54420522015-10-13T22:40:51Z http://ndltd.ncl.edu.tw/handle/04870825052990840915 Applications of machine learning techniques to the design of a self-healing smart grid 機器學習技術於自癒電網的應用 Jui-Jiun Jian 簡瑞均 碩士 國立東華大學 電機工程學系 101 Recently, renewable energy has been promoted all over the world, and the establishment of smart meters and smart grids are gradually becoming a new type of power grid. Although a smart grid gradually forms nowadays, the part of power interruption during transmission still does not have a fine solution. Therefore, the shortage of power may lead customers to be inconvenient, and a huge damage may thus happen in some emergency areas with large demands of power. Besides, the longer the interruption time is, the most serious the damage level of the area without any electricity becomes due to the power interruption. Hence, the repairing scheduling of electrical transmission will decide the degree of damage in these areas. Moreover, most researches about the interruption of the power transmission focused on the methods by switching the original but damaged route to the backup one via circuit breakers. However, if there is no backup route available, customers still face a shortage of electricity. In this work, an optimal repairing scheduling suggestion system for power transmission is designed to decrease the damage for customers and total waiting time for emergency repair work. Besides, we introduce the electrical vehicles (EVs) as the source of the power for emergency situation while repairing, and a Multi-Objective Particle Swarm Optimization (MOPSO) algorithm is utilized to distribute the EVs to decrease the electricity wasted while driving, and avoid the situation that the EVs arrived at the supported place too early or too later as well. The experimental results indicate that the work can effectively lower the time spent in repairing the supported place, and increase the repairing benefit for maintenance crews. Besides, in the part of emergency electricity dispatch for EVs, the proposed method can effectively lower the power shortage for high priority users and the waiting time in repairing for customers. Chenn-Jung Huang 黃振榮 2013 學位論文 ; thesis 62 |
collection |
NDLTD |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 國立東華大學 === 電機工程學系 === 101 === Recently, renewable energy has been promoted all over the world, and the establishment of smart meters and smart grids are gradually becoming a new type of power grid. Although a smart grid gradually forms nowadays, the part of power interruption during transmission still does not have a fine solution. Therefore, the shortage of power may lead customers to be inconvenient, and a huge damage may thus happen in some emergency areas with large demands of power. Besides, the longer the interruption time is, the most serious the damage level of the area without any electricity becomes due to the power interruption. Hence, the repairing scheduling of electrical transmission will decide the degree of damage in these areas. Moreover, most researches about the interruption of the power transmission focused on the methods by switching the original but damaged route to the backup one via circuit breakers. However, if there is no backup route available, customers still face a shortage of electricity.
In this work, an optimal repairing scheduling suggestion system for power transmission is designed to decrease the damage for customers and total waiting time for emergency repair work. Besides, we introduce the electrical vehicles (EVs) as the source of the power for emergency situation while repairing, and a Multi-Objective Particle Swarm Optimization (MOPSO) algorithm is utilized to distribute the EVs to decrease the electricity wasted while driving, and avoid the situation that the EVs arrived at the supported place too early or too later as well.
The experimental results indicate that the work can effectively lower the time spent in repairing the supported place, and increase the repairing benefit for maintenance crews. Besides, in the part of emergency electricity dispatch for EVs, the proposed method can effectively lower the power shortage for high priority users and the waiting time in repairing for customers.
|
author2 |
Chenn-Jung Huang |
author_facet |
Chenn-Jung Huang Jui-Jiun Jian 簡瑞均 |
author |
Jui-Jiun Jian 簡瑞均 |
spellingShingle |
Jui-Jiun Jian 簡瑞均 Applications of machine learning techniques to the design of a self-healing smart grid |
author_sort |
Jui-Jiun Jian |
title |
Applications of machine learning techniques to the design of a self-healing smart grid |
title_short |
Applications of machine learning techniques to the design of a self-healing smart grid |
title_full |
Applications of machine learning techniques to the design of a self-healing smart grid |
title_fullStr |
Applications of machine learning techniques to the design of a self-healing smart grid |
title_full_unstemmed |
Applications of machine learning techniques to the design of a self-healing smart grid |
title_sort |
applications of machine learning techniques to the design of a self-healing smart grid |
publishDate |
2013 |
url |
http://ndltd.ncl.edu.tw/handle/04870825052990840915 |
work_keys_str_mv |
AT juijiunjian applicationsofmachinelearningtechniquestothedesignofaselfhealingsmartgrid AT jiǎnruìjūn applicationsofmachinelearningtechniquestothedesignofaselfhealingsmartgrid AT juijiunjian jīqìxuéxíjìshùyúzìyùdiànwǎngdeyīngyòng AT jiǎnruìjūn jīqìxuéxíjìshùyúzìyùdiànwǎngdeyīngyòng |
_version_ |
1718080235192188928 |