Back to search

Georgia Southern University

Improving Cyber Security in Power Systems Using Neural Networks

Abstract

dc:description.abstract

The modern power grid has become increasingly dependent on wireless communications to fulfill its duties. These new, more automatic, hands-off grids have become cyber-physical power systems. This more recent, more decentralized power grid has created profound changes in how power is generated and distributed. Moreover, to better accommodate these newer technologies inside the older legacy grid, more devices that can connect to or interact with the Internet have been required for sensor and actuator information to be collected, transmitted, and then processed. These devices have created what is known as the Internet of Things (IoT) in the power stations. This creates a more efficient and consistent power grid with less need for human assistance or interference. Of course, the increased utilization of cyber-physical power systems in the modern world has created an increasing concern about what would happen should some person or group stage a cyber-attack on the power grid. In that vein, there is much to identify and quantify the vulnerabilities associated with Supervisory Control and Data Acquisition (SCADA) and Industrial Control System (ICS) devices. It’s still primarily believed that current systems are vulnerable and need improvements as they may suffer from the same issues as the existing systems. This thesis will provide a brief overview of the threats to the power grid and how they are currently being combated. In addition, a potential machine learning neural network-based anomaly detection will be discussed.

Degree

thesis:*
Name thesis:degree_name
Master of Science, Electrical Engineering
Level thesis:degree_level
Thesis (restricted to Georgia Southern)
Discipline thesis:degree_discipline
Department of Electrical Engineering

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Savage, Dylan
Contributors dc:contributor
  • Rocio Alba-Flores
  • Fernando Rios-Gutierrez

Subjects

dc:subject × 8

Rights

Identifiers

dc:identifier.*
Repository record dc:identifier
https://digitalcommons.georgiasouthern.edu/etd/2482

Chain of custody

source
Harvested from
Georgia Southern University
Base URL
digitalcommons.georgiasouthern.edu/do/oai/
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Savage, Dylan. Improving Cyber Security in Power Systems Using Neural Networks. Thesis (restricted to Georgia Southern) thesis, https://digitalcommons.georgiasouthern.edu/etd/2482