Resilient cybersecurity in smart grid ICS communication using BLAKE3-driven dynamic key rotation and intrusion...

Area of Science:

• Cybersecurity
• Industrial Control Systems (ICS)
• Machine Learning

Background:

• Industrial Control Systems (ICS) are increasingly connected to critical infrastructure, creating new cybersecurity vulnerabilities.
• Advanced cyber threats pose significant risks to the security and operational integrity of these systems.

Purpose of the Study:

• To develop an innovative cybersecurity approach for ICS.
• To predict and neutralize diverse cyber attack vectors, including false data injections, denial-of-service, and infiltrations.
• To enhance the security and operational integrity of critical infrastructure.

Main Methods:

• Utilized the IEC 60870-5-104 dataset for analysis.
• Employed Synthetic Minority Oversampling Technique (SMOTE) for data augmentation.
• Implemented a Random Forest Classifier (achieving 1.00 accuracy) and Isolation Forest for anomaly detection.
• Incorporated dynamic AES-256-CBC encryption with BLAKE3-derived keys.
• Applied Bayesian-driven forensic methodology for post-attack analysis.

Main Results:

• Achieved 1.00 accuracy with the Random Forest Classifier.
• Demonstrated cryptographic resilience against ciphertext-only attacks.
• Successfully evaluated against evolving and zero-day attack patterns via real-time simulations.
• Validated model strength through various security tests (e.g., Chi-square, Shannon entropy).

Conclusions:

• The proposed framework effectively predicts and neutralizes ICS cyber threats.
• The hybrid approach enhances operational strength and national security.
• This research addresses critical gaps in Industrial Control System cybersecurity.