Self-adaptive data-driven PID controller for a multi-area interconnected power system under DoS attacks

This article studies the load frequency control (LFC) problem of multi-area interconnected power systems under the threat of denial-of-service (DoS) attacks. Due to the open communication network of control regions, interconnected power systems are particularly vulnerable to Dos attacks, which can undermine the stable operation of the power system. To deal with this problem, we design a novel self-adaptive data-driven proportional-integral-derivative (PID) LFC controller, which is a data-driven control strategy that does not require explicit model information of the interconnected power system. First, a data-driven model which is completely equivalent to the multi-area interconnected power system is constructed by using the dynamic linearization technique, and the radial basis function neural network (RBFNN) is applied to identify the model parameters online. On this basis, the DoS attacks that happen in the output channel are described using Bernoulli distribution, and the real-time compensation mechanism is designed to deal with the DoS attacks adverse effects. Then, an adaptive parameter update algorithm for the PID controller is designed by a rigorous analysis based on the Lyapunov theory such that the tracking error of the system converges asymptotically. In the end, the effectiveness of the developed control algorithm is illustrated by a simulation example of a three-area interconnected power system.