Recursive Unscented Kalman Filtering for Power Distribution Networks Under Hybrid Attacks: Tackling Dynamic Quantization Effects

Abstract

This article investigates the state estimation problem for power distribution networks (PDNs) subject to dynamic quantization effects and hybrid cyberattacks, where measurement signals are transmitted from sensors to a remote filter via open digital communication networks. To enhance bandwidth utilization and ensure reliable data transmission, a dynamic quantization mechanism is introduced, which effectively accommodates the dynamic characteristics of power signals. Furthermore, the system is vulnerable to hybrid cyberattacks that may occur simultaneously in a random manner, including denial-of-service (DoS) attacks and false data injection (FDI) attacks, characterized by Bernoulli distributed random variables. The primary objective of this work is to develop a recursive unscented Kalman filter capable of addressing the combined challenges of measurement nonlinearities, dynamic quantization effects, and hybrid cyberattack scenarios. By solving Riccati-like difference equations, an upper bound on the filtering error covariance is derived and subsequently minimized through the design of time-varying filter gains. Extensive simulations on the IEEE 69 distribution test system demonstrate the effectiveness of the proposed filtering algorithm.