Yang Hanqing, Li Tieshan, Long Yue, Xiao Yang
IEEE Trans Cybern. 2024 Jan;54(1):298-307. doi: 10.1109/TCYB.2022.3218026. Epub 2023 Dec 20.
This article presents an event-triggered distributed secondary control with predictive compensation based on the model-free predictive control under Denial-of-Service (DoS) attacks in ac/dc-networked microgrids. First, models of ac/dc networked microgrids in both electric network and communication network are established. On this premise, event-triggered distributed secondary control is proposed to solve the problems of strong communication burden and low-power distribution accuracy. Besides, aiming at the impact of DoS attacks on distributed secondary control, a compensation algorithm based on model-free predictive control is designed to estimate the control variables when the DoS attack occurs, which can improve the control performance and maintain the stable operation without a specific system structure system. Then, the convergence of event-triggered distributed secondary control with the condition of whether DoS attacks happen are analyzed. Finally, the effectiveness of the proposed control is verified on the hardware-in-loop (HIL) simulation platform consisting of the RT-LAB simulator, MATLAB/Simulink simulation model, and DSP controller.
本文提出了一种在交流/直流联网微电网遭受拒绝服务(DoS)攻击的情况下,基于无模型预测控制的具有预测补偿的事件触发分布式二次控制方法。首先,建立了交流/直流联网微电网在电网和通信网络中的模型。在此前提下,提出事件触发分布式二次控制以解决通信负担重和功率分配精度低的问题。此外,针对DoS攻击对分布式二次控制的影响,设计了一种基于无模型预测控制的补偿算法,用于估计DoS攻击发生时的控制变量,该算法无需特定系统结构即可提高控制性能并维持稳定运行。然后,分析了事件触发分布式二次控制在DoS攻击是否发生情况下的收敛性。最后,在所构建的由RT-LAB模拟器、MATLAB/Simulink仿真模型和DSP控制器组成的硬件在环(HIL)仿真平台上验证了所提控制方法的有效性。