Fast economic dispatch with false data injection attack in electricity-gas cyber-physical system: A data-driven...

Area of Science:

• Cyber-physical systems engineering
• Electrical engineering
• Computer science

Background:

• Electricity-gas cyber-physical systems (EGCPS) are vulnerable to cyberattacks like false data injection attacks (FDIA) due to communication network reliance.
• FDIA poses a significant threat to the secure and efficient operation of EGCPS, impacting economic dispatch.
• Existing methods may lack the accuracy and speed required for real-time FDIA detection and mitigation in complex EGCPS.

Purpose of the Study:

• To develop a data-driven approach for real-time, location-specific FDIA detection in EGCPS.
• To enhance the computational efficiency and cybersecurity of data-driven fast economic dispatch.
• To propose a method for timely recovery of tampered measurements following FDIA.

Main Methods:

• A ResNet-ALSTM model, combining residual networks (ResNet) and attention long short-term memory (ALSTM), was developed for temporal correlation and feature extraction.
• The ResNet-ALSTM model functions as a multi-label classifier to identify outlier locations of tampered measurements.
• The Fast Dynamic Time Warping algorithm was employed for efficient recovery of power system measurements after FDIA.

Main Results:

• The proposed ResNet-ALSTM achieved high accuracy in the locational detection of FDIA within the EGCPS.
• Tampered measurements were effectively recovered using the Fast Dynamic Time Warping algorithm.
• The integrated approach enabled fast economic dispatch in the EGCPS with recovered data, demonstrating high computational efficiency.

Conclusions:

• The developed ResNet-ALSTM model offers a robust solution for safeguarding EGCPS against FDIA.
• The study demonstrates the feasibility of combining advanced machine learning with data recovery for secure and efficient EGCPS operation.
• This research provides a promising pathway for enhancing the resilience of critical energy infrastructure against sophisticated cyber threats.