Zhao, Meng
(2023)
Modernized Power System Optimal Operation & Safety Protection through Mathematical and Artificial Intelligence Techniques.
Doctoral Dissertation, University of Pittsburgh.
(Unpublished)
Abstract
The electricity industry is facing significant expectations and requirements to optimize the energy operation, secure the electrical system and critical infrastructure that depend on electric power against cyber or physical attacks, and maintain system stability. Power and energy optimization problem is one of the most fundamental concerns that must be dealt with in electrical power system management. During the operation of electric power systems, the distribution lines may experience faults caused by unpredictable natural disasters or accidents. Also, Sub-Synchronous Oscillations (SSOs) have recently emerged as a critical concern in modern power systems primarily due to the rapid proliferation of grid-connected wind power plants and installation of series-compensated transmission lines.
To create an economical and optimal power system, this thesis proposes a novel algorithm by leveraging linear approximation and convex relaxation for the non-convex AC Optimal Power Flow (ACOPF) problems, making it effective to get the global near-optimal solutions. To improve the stability and resiliency of power systems, two deep learning models are developed for fault recognition and SSO detection: a real-time deep learning model based on convolution neural network (ConvNet) is developed to classify the fault types and localize the fault with the access to the measured data, and the single observability as its superiority guarantees the performance with limited data access or measurement devices; the second deep learning model integrates ConvNet and long short-term memory (LSTM) techniques for SSO detection and evaluation that harnesses the full potential in pseudo-continuous quadrature wavelet transform (PCQ-WT).
By utilizing mathematical and artificial intelligence techniques, this thesis proposes algorithms for system optimal operation and faulty status detection to guarantee an economical and resilient power system.
Share
| Citation/Export: |
|
| Social Networking: |
|
Details
| Item Type: |
University of Pittsburgh ETD
|
| Status: |
Unpublished |
| Creators/Authors: |
|
| ETD Committee: |
|
| Date: |
13 June 2023 |
| Date Type: |
Publication |
| Defense Date: |
5 April 2023 |
| Approval Date: |
13 June 2023 |
| Submission Date: |
24 February 2023 |
| Access Restriction: |
No restriction; Release the ETD for access worldwide immediately. |
| Number of Pages: |
89 |
| Institution: |
University of Pittsburgh |
| Schools and Programs: |
Swanson School of Engineering > Electrical and Computer Engineering |
| Degree: |
PhD - Doctor of Philosophy |
| Thesis Type: |
Doctoral Dissertation |
| Refereed: |
Yes |
| Uncontrolled Keywords: |
ACOPF, deep learning, fault localization, power system optimization, power system resilience, SSR detection |
| Date Deposited: |
13 Jun 2023 14:11 |
| Last Modified: |
13 Jun 2023 14:11 |
| URI: |
http://d-scholarship.pitt.edu/id/eprint/44232 |
Metrics
Monthly Views for the past 3 years
Plum Analytics
Actions (login required)
 |
View Item |
![[feed]](/images/feed-icon-32x32.png){kind=icon}