Abstract :

Machine Learning (ML)-based Classification Techniques Offer Promising Solutions For Identifying Failures In Intelligent Electronic Devices (IEDs) Within Smart Power Grid Systems. These Systems Are Integral Components Of Modern Power Grids, Facilitating Efficient Energy Management And Ensuring Reliable Electricity Supply. However, The Complexity And Interconnected Nature Of Smart Grid Infrastructures Make Them Susceptible To Various Failures And Attacks, Necessitating Robust Fault Detection Mechanisms. Another Crucial Application Lies In Cybersecurity For Smart Grids, Where ML Algorithms Can Aid In Detecting And Mitigating Attacks Targeting IEDs. By Analyzing Network Traffic Patterns And Abnormal Behaviors In IEDs, ML Models Can Identify Suspicious Activities Indicative Of Failures, Such As Unauthorized Access Attempts Or Tampering With Device Configurations. This Proactive Approach To Cybersecurity Enhances The Resilience Of Smart Grid Systems Against Malicious Threats, Safeguarding Critical Infrastructure And Ensuring Uninterrupted Electricity Supply To Consumers. Current Methods For Detecting IED Failures In Smart Power Grid Systems Often Rely On Rule-based Approaches Or Manual Inspection, Which Are Labor-intensive And Prone To Errors. These Traditional Techniques May Overlook Subtle Patterns Or Anomalies Indicative Of Emerging Failures, Leading To Delayed Responses And Increased Risk Of System Downtime. Additionally, Existing Fault Detection Mechanisms May Struggle To Differentiate Between Genuine Failures And Benign Fluctuations In System Behavior, Resulting In False Alarms And Unnecessary Maintenance Interventions. To Address The Limitations Of Existing Fault Detection Methods, This Work Proposes A Novel ML-based Classification System For Identifying Failures In IEDs Within Smart Power Grid Systems. The Proposed System Leverages Supervised Learning Algorithms Trained On Labeled Datasets Derived From Power System Attack Scenarios. By Analyzing Various Features E

Published:

09-6-2025

Issue:

Vol. 25 No. 6 (2025)

Page Nos:

96-109

Section:

Articles

License:

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

How to Cite