Energy efficient cyber-physical control of renewable microgrids using edge-AI enabled IoT and secure blockchain...

Area of Science:

• Electrical Engineering
• Computer Science
• Cybersecurity

Background:

• Increasing microgrid complexity necessitates secure and efficient cyber-physical coordination.
• Renewable energy integration introduces challenges in managing distributed energy resources.

Purpose of the Study:

• To design and evaluate an edge-AI blockchain framework for cyber-physical management of renewable-integrated microgrids.
• To enhance security, efficiency, and resilience in microgrid operations.

Main Methods:

• Utilized Spiking Neural Networks (SNNs) on edge devices for real-time learning and fault detection.
• Implemented Hyperledger Fabric for energy transaction validation and access control.
• Integrated edge AI with blockchain for secure peer-to-peer communication and data provenance.

Main Results:

• Achieved 97.6% Cyber Fault Detection Accuracy (CFDA).
• Demonstrated Consensus Delay < 2.3 s and Voltage Deviation < ±1.1%.
• Reduced communication overhead and improved energy authentication efficiency by 28% compared to centralized control.

Conclusions:

• The edge-AI blockchain framework provides enhanced cyber resilience and real-time suitability for microgrids.
• The system improves stability, fault recovery, and controlability, aligning with decentralized microgrid standards.
• Edge AI ensures anomaly detection with minimal computation, maintaining grid observability.