Switching System-Based Event-Triggered Sampled-Data Control for Multi-Area Power Systems Under DoS Attacks: A Data-Driven Method

The reliance of modern power systems on open communication networks for load frequency control (LFC) increases vulnerability to cyber-attacks, particularly denial-of-service (DoS) attacks. This study explores a novel switching system-based resilient event-triggered sampled-data control for LFC using a data-driven method. Firstly, a linear data-dependent model for power systems is presented addressing unknown nonlinearities, treating sampling period as a vital design parameter. Further, the LFC system is modeled as multiple switching subsystems under DoS attacks, and then a new adaptive sampled-data controller is proposed to accomplish frequency regulation while withstanding the attacks. Additionally, an event-triggered scheduling mechanism (ETSM) is considered to optimize bandwidth resources for data transmission during sampling. To assess the proposed LFC scheme, a new stability criterion and analysis method, employing the average dwell time (ADT) technique and Lyapunov stability theory, is introduced. Finally, the scheme's effectiveness is demonstrated through evaluation on a three-area power system.