Dissertation/Thesis Abstract

Real-Time Stability Surveillance in Power Systems: A Deep Learning Approach
by Shinde, Bhavesh Tukaram, M.S., The George Washington University, 2020, 96; 27669739
Abstract (Summary)

Online Power System Stability assessment is a critical problem which has enormous scope of development. Most electrical utilities investigate system stability by simulating critical contingencies to determine the severity of transient disturbances in the system. Assessment of power system transient stability is critical for a reliable and continuous operation and to ensure none of the working generating units in the system go out of synchronism. The main objective of this research is to develop a fast and robust online transient stability assessment tool to classify the system operating states and to identify system critical generators in case of instability. This research proposes a deep learning neural network framework that captures the phasor measurement unit (PMU) measurements and monitor the system transient stability in real-time. The proposed framework in a first case study utilizes the convolutional neural network (CNN) with hypotheses CNN pooling (HCP) to identify the system operating states and detect the set of critical generators. The proposed framework in the second case utilizes a hybrid deep learning network consisting of CNN and Long Short Term Memory (LSTM) called ConvLSTM network for the given problem of system stability monitoring. The suggested CNN-HCP module and ConvLSTM module for stability assessment and for detecting critical generators through multi-class and multi-label classifications are tested on the IEEE 118-bus test system and IEEE 39-bus test system, respectively, where different types of faults at different locations and under varying system load conditions are simulated. The test results verified that our proposed framework is fast and accurate, thereby a viable approach for online system stability monitoring applications.

Indexing (document details)
Advisor: Dehghanian, Payman
Commitee: Ahmadi, Shahrokh, Doroslovacki, Milos
School: The George Washington University
Department: Electrical Engineering
School Location: United States -- District of Columbia
Source: MAI 81/6(E), Masters Abstracts International
Source Type: DISSERTATION
Subjects: Electrical engineering, Engineering
Keywords: Convolutional long-short term memory (ConvLSTM), Hypotheses CNN pooling (HCP), Phasor measurement unit (PMU), Power system stability, Power system stability monitoring, Transient Stability Analysis (TSA)
Publication Number: 27669739
ISBN: 9781392520536
Copyright © 2020 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy
ProQuest