Real-Time Intrusion Detection in Power Grids Using Deep Learning: Ensuring DPU Data Security

Deep learning technologies have revolutionized the management of energy, energy consumption, and data security within smart grids through non-intrusive load monitoring (NILM). This paper explores the use of deep learning for real-time intrusion detection in power grids with a primary focus on safegu...

Full description

Saved in:
Bibliographic Details
Main Authors: Maoran Xiao, Qi Zhou, Zhen Zhang, Junjie Yin
Format: Article
Language:English
Published: Ital Publication 2024-09-01
Series:HighTech and Innovation Journal
Subjects:
Online Access:https://hightechjournal.org/index.php/HIJ/article/view/766
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850284526201208832
author Maoran Xiao
Qi Zhou
Zhen Zhang
Junjie Yin
author_facet Maoran Xiao
Qi Zhou
Zhen Zhang
Junjie Yin
author_sort Maoran Xiao
collection DOAJ
description Deep learning technologies have revolutionized the management of energy, energy consumption, and data security within smart grids through non-intrusive load monitoring (NILM). This paper explores the use of deep learning for real-time intrusion detection in power grids with a primary focus on safeguarding the integrity and security of Data Processing Units (DPUs). An evaluation of various machine learning models, including Support Vector Machine (SVM), Linear Discriminant Analysis (LDA), Decision Trees, and Random Forests, is conducted to detect various types of intrusions, including Fault, Injection, Masquerade, Normal, and Replay. Random Forest produced AUC values of 1.00 for all classes and an overall F1-score of 0.99 for all classes. The Decision Tree model also shows robust performance for detecting Fault and Injection intrusions (AUC = 0.98), with an overall F1-score of 0.94. However, the LDA and SVM models do not perform well in detecting Injection intrusions with overall F1-scores of 0.83 and 0.86. Advances in machine learning can be used to improve smart grid security, reliability, and efficiency, according to this study. These findings highlight the potential of advanced machine learning techniques to enhance smart grid reliability and efficiency.   Doi: 10.28991/HIJ-2024-05-03-018 Full Text: PDF
format Article
id doaj-art-1d4adca539c74fad95ddbf136f797208
institution OA Journals
issn 2723-9535
language English
publishDate 2024-09-01
publisher Ital Publication
record_format Article
series HighTech and Innovation Journal
spelling doaj-art-1d4adca539c74fad95ddbf136f7972082025-08-20T01:47:33ZengItal PublicationHighTech and Innovation Journal2723-95352024-09-015381482710.28991/HIJ-2024-05-03-018214Real-Time Intrusion Detection in Power Grids Using Deep Learning: Ensuring DPU Data SecurityMaoran Xiao0Qi Zhou1Zhen Zhang2Junjie Yin31) State Grid Jiangsu Electric Power Co., Ltd. Limited Information and Telecommunication Branch, Nanjing, Jiangsu, 210000, China. 2) State Grid Jiangsu Electric Power Co., Ltd. Wuxi Power Supply Branch, Wuxi, Jiangsu, 214000, China.State Grid Jiangsu Electric Power Co., Ltd. Wuxi Power Supply Branch, Wuxi, Jiangsu, 214000,State Grid Jiangsu Electric Power Co., Ltd. Limited Information and Telecommunication Branch, Nanjing, Jiangsu, 210000,State Grid Jiangsu Electric Power Co., Ltd. Limited Information and Telecommunication Branch, Nanjing, Jiangsu, 210000,Deep learning technologies have revolutionized the management of energy, energy consumption, and data security within smart grids through non-intrusive load monitoring (NILM). This paper explores the use of deep learning for real-time intrusion detection in power grids with a primary focus on safeguarding the integrity and security of Data Processing Units (DPUs). An evaluation of various machine learning models, including Support Vector Machine (SVM), Linear Discriminant Analysis (LDA), Decision Trees, and Random Forests, is conducted to detect various types of intrusions, including Fault, Injection, Masquerade, Normal, and Replay. Random Forest produced AUC values of 1.00 for all classes and an overall F1-score of 0.99 for all classes. The Decision Tree model also shows robust performance for detecting Fault and Injection intrusions (AUC = 0.98), with an overall F1-score of 0.94. However, the LDA and SVM models do not perform well in detecting Injection intrusions with overall F1-scores of 0.83 and 0.86. Advances in machine learning can be used to improve smart grid security, reliability, and efficiency, according to this study. These findings highlight the potential of advanced machine learning techniques to enhance smart grid reliability and efficiency.   Doi: 10.28991/HIJ-2024-05-03-018 Full Text: PDFhttps://hightechjournal.org/index.php/HIJ/article/view/766machine learningintrusion detectionsmart gridsdata integritysecuritynilmreal-time detectionenergy management.
spellingShingle Maoran Xiao
Qi Zhou
Zhen Zhang
Junjie Yin
Real-Time Intrusion Detection in Power Grids Using Deep Learning: Ensuring DPU Data Security
HighTech and Innovation Journal
machine learning
intrusion detection
smart grids
data integrity
security
nilm
real-time detection
energy management.
title Real-Time Intrusion Detection in Power Grids Using Deep Learning: Ensuring DPU Data Security
title_full Real-Time Intrusion Detection in Power Grids Using Deep Learning: Ensuring DPU Data Security
title_fullStr Real-Time Intrusion Detection in Power Grids Using Deep Learning: Ensuring DPU Data Security
title_full_unstemmed Real-Time Intrusion Detection in Power Grids Using Deep Learning: Ensuring DPU Data Security
title_short Real-Time Intrusion Detection in Power Grids Using Deep Learning: Ensuring DPU Data Security
title_sort real time intrusion detection in power grids using deep learning ensuring dpu data security
topic machine learning
intrusion detection
smart grids
data integrity
security
nilm
real-time detection
energy management.
url https://hightechjournal.org/index.php/HIJ/article/view/766
work_keys_str_mv AT maoranxiao realtimeintrusiondetectioninpowergridsusingdeeplearningensuringdpudatasecurity
AT qizhou realtimeintrusiondetectioninpowergridsusingdeeplearningensuringdpudatasecurity
AT zhenzhang realtimeintrusiondetectioninpowergridsusingdeeplearningensuringdpudatasecurity
AT junjieyin realtimeintrusiondetectioninpowergridsusingdeeplearningensuringdpudatasecurity