Electrical and Computer Engineering ETDs

Publication Date

Fall 12-17-2022

Abstract

Modernizing power grids with communication-based technologies has introduced new challenges to the operation of the grid, especially when the communication network experiences failure or data is corrupted during the transfer. This issue is studied in this dissertation from the control and protection perspective. First, the cyber attack detection and mitigation of distributed control of microgrids is addressed when the distributed energy resources (DER) are exposed to false data injection attacks. A cyber-threat detection technique is proposed based on Kullback-Liebler divergence-based criterion. This criterion with a threshold can detect the misbehavior of a compromised DER control unit and, consequently, calculates the interior-belief factor and communicates it with its neighboring DERs to inform them of the reliability of its outgoing information. From the protection point of view, the focus is on adaptive protection systems' communication failure, which relies highly on the communication network. To address the communication failure of the adaptive protection system, this dissertation proposes a local adaptive modular protection (LAMP) architecture. A LAMP unit provides adaptive, setting-less, and communication-free protection. Using the locally measured information of voltages and currents, this module can find the circuit topology using machine learning techniques without any information from other parts of the distribution system. After finding the topology, each LAMP uses machine learning algorithms for identifying fault types. Moreover, proper coordination between relays is necessary when a fault occurs in the system. This can be achieved using zone classification. Machine learning algorithms are used during zone classification to find the fault zone and location. This dissertation has verified the validity of the proposed approaches using the simulation of different test circuits in various commercially available software packages.

Keywords

Distributed Energy Resources, Microgrid, Cyber Attack, Machine Learning, Support Vector Machine, Short Circuit

Document Type

Dissertation

Language

English

Degree Name

Electrical Engineering

Level of Degree

Doctoral

Department Name

Electrical and Computer Engineering

First Committee Member (Chair)

Ali Bidram

Second Committee Member

Michael Devetsikiotis

Third Committee Member

Matthew J. Reno

Fourth Committee Member

Reinaldo Tonkoski

Download

Share

COinS