Ensemble Model Based on Hybrid Deep Learning for Intrusion Detection in Smart Grid Networks

Area of Science:

• Electrical Engineering
• Computer Science
• Cybersecurity

Background:

• Smart Grid systems enhance electric grid efficiency and reliability through digital technologies.
• Communication networks are vital but introduce vulnerabilities to cyberattacks, risking grid stability.
• Intrusion detection and prevention are critical for mitigating these cyber threats.

Purpose of the Study:

• To propose a hybrid deep-learning approach for detecting distributed denial-of-service (DDoS) attacks.
• To enhance the security and resilience of Smart Grid communication infrastructure.
• To develop a real-time monitoring system for attack surveillance.

Main Methods:

• A hybrid deep-learning model combining Convolutional Neural Network (CNN) and Recurrent Gated Unit (GRU) algorithms.
• Utilized two datasets: Canadian Institute for Cybersecurity's Intrusion Detection System dataset and a custom Omnet++ simulated dataset.
• Developed a Kafka-based dashboard for real-time monitoring and attack surveillance.

Main Results:

• The proposed hybrid deep-learning model achieved a high detection accuracy of 99.86%.
• Demonstrated effective detection of distributed denial-of-service attacks in simulated and real-world datasets.
• The real-time monitoring dashboard facilitated efficient attack surveillance.

Conclusions:

• The hybrid deep-learning approach is highly effective for detecting cyberattacks in Smart Grid communication networks.
• This method significantly improves the security posture and reliability of the Smart Grid.
• The developed system offers a robust solution for real-time threat detection and mitigation.