Author: Zhou, Jiaqi
Title: TSPM-CNN based power system transient stability analysis using a new data extraction method
Advisors: Bu, S. Q. (EEE)
Degree: M.Sc.
Year: 2022
Subject: Electric power system stability
Electric power systems -- Data processing
Hong Kong Polytechnic University -- Dissertations
Department: Department of Electrical Engineering
Pages: xii, 119 pages : color illustrations
Language: English
Abstract: In the future, the power system may operate closer to the stability limit in order to improve its efficiency and economic value. The access of a large number of new energy sources and the reform of power marketization make the system dynamic changes more complex, and the traditional model-based power system stability assessment methods are no longer able to meet the requirements of online grid dispatch. Meanwhile, the increasing number of Wide-Area Measurement System (WAMS) and Phasor Measurement Unit (PMU) configurations in power grids has effectively improved the situational awareness of the system, and coupled with the increasing computer computing power, the data-driven transient stability assessment (TSA) method based on the data has been widely studied.
Although such methods have many advantages, there is still room for improvement in many aspects such as model selection and feature construction. To address the shortcomings of the existing methods and also to further improve the speed and accuracy of prediction, this paper conducts research on transient stability prediction methods for power systems based on deep learning theory, and the main results of the paper are as follows.
In order to improve the accuracy of transient stability prediction, a new data extraction method based on deep learning TSA is proposed in this paper. And the proposed new data extraction method is evaluated by selecting suitable evaluation indexes with the New England IEEE-39 bus system as the test system.
The method makes use of the excellent feature self-extraction capability of convolutional neural network (CNN), makes full use of the timing information of individual features and the correlation information of different features, effectively solves the problems of insufficient data extraction and data timing information loss in the traditional data-driven power system transient stability prediction method, and obtains more accurate power system transient stability prediction than the traditional two-dimensional data extraction.
In addition, we find that the length of extracted data is proportional to the model prediction accuracy but inversely proportional to the model prediction time in practical applications. Therefore, in order to solve the problem that the length of data is inversely proportional to the prediction time, this paper proposes a transient stability prediction method based on the combination of time-series prediction network and CNN.
The method utilizes the features of the time-series prediction model (TSPM) to predict the subsequent data of the power system PMU and outputs the transient stability prediction results of the power system in combination with the CNN. The method makes full use of the temporal prediction capability of the temporal prediction network and the feature self-extraction capability of the CNN to achieve fast and highly accurate power system transient stability prediction with less data requirements.
Rights: All rights reserved
Access: restricted access

Files in This Item:
File Description SizeFormat 
6955.pdfFor All Users (off-campus access for PolyU Staff & Students only)3.82 MBAdobe PDFView/Open

Copyright Undertaking

As a bona fide Library user, I declare that:

  1. I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
  2. I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
  3. I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.

By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.

Show full item record

Please use this identifier to cite or link to this item: