Ajuz, Ashley
(2024)
Detecting Energy Theft Cyberattacks in Smart Metering Infrastructure using Deep Learning Model and Rule-Based Policy.
Master's Thesis, University of Pittsburgh.
(Unpublished)
Abstract
The Advanced Metering Infrastructure (AMI) is an integral part of smart energy grid development and provides a bi-directional communication system between utilities and con-sumers. AMI offers several benefits such as automated collection of time-series energy measurements, outage management, connection and disconnection requests, and system status tracking. However, given the increased dependency on device connectivity and data-driven
decisions, AMI and smart energy systems become vulnerable to multiple security threats. This work will focus on energy theft, which is a well-known data-falsification cyberattack
on AMI. In this attack, smart meters are compromised to report falsified energy consumption data, leading to financial losses, potential disruption of power flow, and other negative impacts on the functionality of power grid components. In this thesis, we propose an energy
theft attack detection approach utilizing deep machine learning (ML) algorithm, specifically
a long-short term memory recurrent neural network (LSTM-RNN) model. The proposed approach combines an LSTM-RNN deep learning model and rule-based policy for attack
detection. Unlike prior work that mainly considers supervised machine learning models, an unsupervised learning approach is proposed that would be more practical in real-world scenarios. The performance of both supervised and unsupervised learning-based approaches is compared to show the effectiveness of the proposed theft detection method under different attack models. Various tests are conducted to evaluate the attack detection performance using a publicly available dataset containing energy consumption readings. It is shown that the proposed model achieves high attack detection accuracy and low false alarm rates.
Share
| Citation/Export: |
|
| Social Networking: |
|
Details
| Item Type: |
University of Pittsburgh ETD
|
| Status: |
Unpublished |
| Creators/Authors: |
|
| ETD Committee: |
|
| Date: |
6 September 2024 |
| Date Type: |
Publication |
| Defense Date: |
15 July 2024 |
| Approval Date: |
6 September 2024 |
| Submission Date: |
26 July 2024 |
| Access Restriction: |
No restriction; Release the ETD for access worldwide immediately. |
| Number of Pages: |
56 |
| Institution: |
University of Pittsburgh |
| Schools and Programs: |
Swanson School of Engineering > Electrical and Computer Engineering |
| Degree: |
MS - Master of Science |
| Thesis Type: |
Master's Thesis |
| Refereed: |
Yes |
| Uncontrolled Keywords: |
energy theft detection, cyberattacks, AMI, LSTM, deep machine learning. |
| Date Deposited: |
06 Sep 2024 20:05 |
| Last Modified: |
06 Sep 2024 20:05 |
| URI: |
http://d-scholarship.pitt.edu/id/eprint/46762 |
Metrics
Monthly Views for the past 3 years
Plum Analytics
Actions (login required)
 |
View Item |
![[feed]](/images/feed-icon-32x32.png){kind=icon}