Enhancing resilience of distributed DC microgrids against cyber attacks using a transformer-based Kalman filter...

Area of Science:

• Electrical Engineering
• Cybersecurity
• Control Systems

Background:

• Distributed DC microgrids are vulnerable to cyber attacks like FDI, DoS, and delay attacks.
• Existing methods struggle with noisy data and high-dimensional data extraction in microgrid operations.

Purpose of the Study:

• To develop a data-driven methodology for enhancing the resilience of distributed DC microgrids against cyber attacks.
• To improve the accuracy of signal transmission prediction in noisy environments.

Main Methods:

• Developed a Transformer-based Kalman Filter (TKF) estimator.
• Integrated an AutoRegressive Integrated Moving Average (ARIMA) model for state-space representation.
• Utilized deep learning, combining transformers and Long Short-Term Memory (LSTM) networks for data extraction.

Main Results:

• Demonstrated the efficacy of the TKF estimator in maintaining microgrid stability under various attack scenarios through simulations.
• Achieved significant improvements in estimation accuracy and system performance.
• Validated the robustness of the proposed method against cyber threats.

Conclusions:

• The proposed TKF methodology effectively enhances DC microgrid resilience against cyber attacks.
• The integration of deep learning and advanced filtering techniques shows promise for future microgrid security.
• Further research can explore advanced filtering and deep learning for improved adaptability to nonlinearities.