Abstract :

The Growing Integration Of Intelligent Cyber-physical Power Systems With Communication Networks Has Heightened Vulnerability To Cyber Attacks, Necessitating Sophisticated Intrusion Detection Technologies. This Study Introduces A Deep Learning Architecture Utilizing The Cybersecurity Intrusion Simulated Network Dataset And PSCADgenerated Cyber Threat Scenarios. Data Preprocessing Encompasses Normalization, Category Encoding, And SMOTEENN Sampling. Various Models, Specifically Convolutional Neural Network, Long Short Term Memory, Transformer, And A Hybrid CNN LSTM Architecture, Are Trained Utilizing Hyperparameter Optimization. Experimental Assessment Indicates That The CNN LSTM Hybrid Attains Superior Performance, With 95.3% Accuracy And 94.4% F1 Score On The Cyber Threat Dataset, And 99.9% Accuracy With A 99.9% F1 Score On PSCAD Simulations. Explainable AI Methodologies, Such As LIME And SHAP, Are Employed To Elucidate Feature Contributions And Bolster Trust. The Optimized Model Is Deployed Using A Flask-based Web Application, Facilitating Real-time Monitoring. The System Categorizes Grid Traffic Into The Following Classifications: No Attack, Assault Identified, Injection Attack, Man-in-themiddle (MITM) Attack, Replay Attack, And Spoofing Attack. The Suggested Method Provides Precise, Comprehensible, And Scalable Intrusion Detection For Robust Smart Grid Cybersecurity Operations In Global Implementations.

Published:

08-4-2026

Issue:

Vol. 26 No. 4 (2026)

Page Nos:

1757-1767

Section:

Articles

License:

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

How to Cite