Abstract

In recent years, with the expansion of power system size, the increase of interconnection and the use of large-scale renewable energy, power system stability and safe operations have become more prominent, causing challenges to the normal operation of power grid. Traditional analysis rely on detailed models of the system. But for real power systems, the operating state of the system is variable, and the model-based analysis methods may not accurately reflect the real operating state of the system. Therefore, this dissertation is focused on data-driven stability analysis and control. First, a method for locating the oscillation source of multi-machine systems is proposed. The electromagnetic torque expressions of various generators in a multi-machine system are deduced, and it is found that in each oscillation mode, the electromagnetic torque can be decomposed into a damping torque and a synchronous torque. Based on this development, an oscillation source positioning scheme based on decoupling mode is proposed. Then, a transfer and CNN-LSTM-based method is developed to accelerate and improve the accuracy of the dynamic frequency prediction process. The proposed method exploits system spatial-temporal information and mines the local features of inputs, which highly improves the performance compared with other machine learning methods. Next, a Distributional Soft Actor-Critic (DSAC) method is developed to solve the emergency frequency control problem. The frequency control is formulated as a MDP problem and solved through a novel distributional deep reinforcement learning method. Further, high penetration renewable energy source increase the system uncertainties and impact the cyber security. We propose a detection method based on Bayesian GAN. It can successfully distinguish between securely operating measurements and those that have been attacked with imbalanced training data. Simulation results of this dissertation show the effectiveness of the proposed methods.

Notes

If this is your thesis or dissertation, and want to learn how to access it or for more information about readership statistics, contact us at STARS@ucf.edu

Graduation Date

2022

Semester

Fall

Advisor

Sun, Wei

Degree

Doctor of Philosophy (Ph.D.)

College

College of Engineering and Computer Science

Department

Electrical and Computer Engineering

Degree Program

Electrical Engineering

Identifier

CFE0009843; DP0027784

URL

https://purls.library.ucf.edu/go/DP0027784

Language

English

Release Date

June 2023

Length of Campus-only Access

None

Access Status

Doctoral Dissertation (Open Access)

Share

COinS