Attack estimation–based resilient control for cyber-physical power systems

• Published in: Transactions of the Institute of Measurement and Control 2023-03, Vol.45 (5), p.886-898
• Main Authors: Pan, Kunpeng, Yang, Feisheng, Feng, Zhaowen, Pan, Quan
• Format: Article
• Language: English
• Subjects: Control systems; Cyber-physical systems; H-infinity control; Machine learning; Output feedback; Performance indices; Robust control; State feedback
• ISSN: 0142-3312; 1477-0369
• DOI: 10.1177/01423312221122471

This article is concerned with the resilient control problem for cyber-physical power systems. The main difference from the existing works lies in that the proposed resilient control scheme focuses on robust control in consideration of the attack reconstruction. First, power systems under false data injection attacks are modeled as cyber-physical power systems through the characteristics of power systems and attacks. Second, we design a robust adaptive attack estimation observer with a prescribed H ∞ performance index to reconstruct the attack signals. Sufficient conditions for the design of the observer are formulated into matrix inequalities by choosing an appropriate adaptive law for online learning. Third, on the basis of attack estimation, a resilient control strategy by dynamic output feedback is proposed, which avoids the difficulty in designing state feedback resilient control based on the observer. Meanwhile, the attack estimation observer and output feedback resilient control are designed separately, and their performance is considered respectively, which simplifies the design process. Finally, a power system with three generators and six buses is used as an example to verify the feasibility of attack estimation and resilient control scheme.