東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

FindBook

Google Book

Amazon

博客來

Learning to Operate a Sustainable Power System.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Learning to Operate a Sustainable Power System./
作者:	Chen, Yize.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2021,
面頁冊數:	118 p.
附註:	Source: Dissertations Abstracts International, Volume: 83-02, Section: B.
Contained By:	Dissertations Abstracts International83-02B.
標題:	Electrical engineering. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28543926
ISBN:	9798535503516

Learning to Operate a Sustainable Power System.
Chen, Yize.

Learning to Operate a Sustainable Power System. - Ann Arbor : ProQuest Dissertations & Theses, 2021 - 118 p.

Source: Dissertations Abstracts International, Volume: 83-02, Section: B.

Thesis (Ph.D.)--University of Washington, 2021.

This item must not be sold to any third party vendors.

The electric power system is undergoing dramatic transformations due to the emergence of renewable resources and demand-side revolutions. However, in order to face the increasing level of system complexity and uncertainty, we need to come up with algorithms that are able to operate the power grid in a safe, reliable and sustainable manner. Such algorithms need to be compatible with infrastructure advancements as well as societal needs. We argue in this thesis that data-driven methods combined with physical knowledge and optimization theories provide a set of powerful tools that can achieve performance guarantees in a diverse set of energy system applications.This dissertation covers the uncertainty modeling, optimal control, and decision-making for power systems with high penetration of renewables. Leveraging the availability of sensing, actuation and data platforms, we are able to construct specifically-design data-driven algorithms by incorporating domain knowledge such as time series characterization, function convexity, and optimization problem structures. In particular, we present three algorithmic contributions in this work: we present a generative model to forecast possible future scenarios of renewable generators; we illustrate an optimal control framework which is built upon on input convex neural network; we formulate a machine learning task for the optimal power flow problem, and we come up with a generalizable and robust learning-based optimization solver to enable real-time decision-making under the environment of stochastic electricity demand. The development of such algorithms is built upon the foundation of control, optimization and machine learning theories. In addition to these algorithmic contributions, a central focus of this thesis is the application of such data-driven tools in many challenging energy system tasks, while extending the state of the art. The developed methods have enabled more robust electric grid planning and energy storage operations. We also apply the neural network controller to voltage regulation and building energy management tasks, where in both domains the underlying systems are hard to model, and our approaches demonstrate the possibility of achieving optimal control with only sensing data available.

ISBN: 9798535503516Subjects--Topical Terms:

649834
Electrical engineering.
Subjects--Index Terms:

Control

Learning to Operate a Sustainable Power System.
LDR:03508nmm a2200433 4500 001 2342366
005 20220318093124.5
008 241004s2021 ||||||||||||||||| ||eng d
020 $a 9798535503516
035 $a (MiAaPQ)AAI28543926
035 $a AAI28543926
040 $a MiAaPQ $c MiAaPQ
100 1 $a Chen, Yize. $3 3680717
245 1 0 $a Learning to Operate a Sustainable Power System.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2021
300 $a 118 p.
500 $a Source: Dissertations Abstracts International, Volume: 83-02, Section: B.
500 $a Advisor: Zhang, Baosen.
502 $a Thesis (Ph.D.)--University of Washington, 2021.
506 $a This item must not be sold to any third party vendors.
520 $a The electric power system is undergoing dramatic transformations due to the emergence of renewable resources and demand-side revolutions. However, in order to face the increasing level of system complexity and uncertainty, we need to come up with algorithms that are able to operate the power grid in a safe, reliable and sustainable manner. Such algorithms need to be compatible with infrastructure advancements as well as societal needs. We argue in this thesis that data-driven methods combined with physical knowledge and optimization theories provide a set of powerful tools that can achieve performance guarantees in a diverse set of energy system applications.This dissertation covers the uncertainty modeling, optimal control, and decision-making for power systems with high penetration of renewables. Leveraging the availability of sensing, actuation and data platforms, we are able to construct specifically-design data-driven algorithms by incorporating domain knowledge such as time series characterization, function convexity, and optimization problem structures. In particular, we present three algorithmic contributions in this work: we present a generative model to forecast possible future scenarios of renewable generators; we illustrate an optimal control framework which is built upon on input convex neural network; we formulate a machine learning task for the optimal power flow problem, and we come up with a generalizable and robust learning-based optimization solver to enable real-time decision-making under the environment of stochastic electricity demand. The development of such algorithms is built upon the foundation of control, optimization and machine learning theories. In addition to these algorithmic contributions, a central focus of this thesis is the application of such data-driven tools in many challenging energy system tasks, while extending the state of the art. The developed methods have enabled more robust electric grid planning and energy storage operations. We also apply the neural network controller to voltage regulation and building energy management tasks, where in both domains the underlying systems are hard to model, and our approaches demonstrate the possibility of achieving optimal control with only sensing data available.
590 $a School code: 0250.
650 4 $a Electrical engineering. $3 649834
650 4 $a Energy. $3 876794
650 4 $a Computer science. $3 523869
650 4 $a Sustainability. $3 1029978
650 4 $a Control theory. $3 535880
650 4 $a Research. $3 531893
650 4 $a Collaboration. $3 3556296
650 4 $a Electricity distribution. $3 3562889
650 4 $a Buildings. $3 861672
650 4 $a Clean technology. $3 3564885
650 4 $a Optimization. $3 891104
650 4 $a Dissertations & theses. $3 3560115
650 4 $a Climate change. $2 bicssc $3 2079509
650 4 $a COVID-19. $3 3554449
650 4 $a Physics. $3 516296
650 4 $a Control algorithms. $3 3560702
650 4 $a Wind power. $3 672558
650 4 $a Neural networks. $3 677449
650 4 $a Energy management. $3 3680718
650 4 $a Design. $3 518875
650 4 $a Engineering. $3 586835
650 4 $a Methods. $3 3560391
650 4 $a Alternative energy sources. $3 3561089
653 $a Control
653 $a Machine Learning
653 $a Optimization
653 $a Power systems
653 $a Electrical power
690 $a 0544
690 $a 0791
690 $a 0984
690 $a 0640
690 $a 0389
690 $a 0404
690 $a 0605
690 $a 0537
710 2 $a University of Washington. $b Electrical and Computer Engineering. $3 3437797
773 0 $t Dissertations Abstracts International $g 83-02B.
790 $a 0250
791 $a Ph.D.
792 $a 2021
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28543926