Resilient Predictive Load Frequency Control of Multiarea Interconnected Power Systems With Privacy Preserving and Active Detection Against Stealthy Cyber Attacks

Aiming to address the stealthy cyber attacks faced by multiarea interconnected power systems, this article proposes a new decentralized resilient predictive load frequency control (LFC) scheme, which has several prominent features, such as privacy preservation, active attack detection, network blocking defense, and attack tolerance. Compared with the relevant studies in the literature, the novelty of these features is specifically demonstrated by: 1) a dynamic scaling and masking method is constructed to achieve the real-time dynamic protection of network transmission data, which not only enables a closed-loop cyber privacy preservation against eavesdropping attacks, but also guarantees the effectiveness of active detection of attack appearance and disappearance; 2) a scaled one-step-ahead predictive interpolation control strategy is further proposed to achieve dynamic privacy preservation of control algorithm, and to apply a check-before-use mechanism to avoid the false data injected by the stealthy attacks to be loaded into the actuator; and 3) an active attack defense method with fail-safe feature is constructed to block the attacks from penetrating the actuator from the input transmission channels, while providing some finite-time open-loop control capability to suboptimally regulate the LFC before the attacks disappear. A case study of a three-area power system is given to validate the effectiveness of the proposed LFC method.