A Distributed Projection Operator-Based Unknown Input Observer for Attack Estimation and Mitigation on DC Microgrids.

Area of Science:

• Electrical Engineering
• Computer Science
• Control Systems

Background:

• Modern microgrid (MG) systems face critical cybersecurity threats from signal manipulation.
• False data injection attacks pose a significant risk to the safe and reliable operation of DC microgrids.
• Existing cybersecurity measures may require secure channels or additional data, limiting their practicality.

Purpose of the Study:

• To develop a novel model-based observer system for detecting cyberattacks in DC MGs.
• To investigate the effectiveness of a projection operator (PO)-based unknown input observer against false data injection attacks.
• To propose a method for reconstructing and mitigating cyberattacks without secure communication channels.

Main Methods:

• A bank of PO-based unknown input observers is designed for each distributed generation unit.
• Each observer estimates the dynamic states of neighboring units.
• Cyberattack reconstruction is performed by comparing estimated and measured states to detect anomalies.

Main Results:

• The proposed PO-based observer system effectively detects false data injection attacks on transmitted signals.
• The method successfully reconstructs detected attack signals to restore data integrity.
• Cyberattacks are mitigated, ensuring the reliable operation of DC MGs.
• The system demonstrated practical efficiency and advantages over existing methods through real-time simulations.

Conclusions:

• The novel PO-based observer system offers a robust solution for DC MG cybersecurity against false data injection attacks.
• This approach enhances microgrid resilience by enabling effective attack detection and mitigation.
• The method's independence from secure channels or additional information transfer makes it a practical and advantageous solution.