False Data Injection Attacks Against State-of-Charge Estimation of Battery Energy Storage Systems in Smart Distribution Networks

Abstract

Advancement of battery energy storage systems (BESSs) has made BESSs typical cyber-physical systems (CPSs), which exposes BESSs, especially for the state of charge (SoC) estimation function, to severe cyber attacks. This article investigates the vulnerability of SoC estimation of BESSs in smart distribution networks (SDNs) to false data injection attacks (FDIAs) to provide a basis to study the attack mechanisms against BESSs and a theoretical guide to develop countermeasures. The mechanism of FDIAs against SoC estimation is investigated by theoretically analyzing the SoC estimation errors caused by compromised measurements, and a static FDIA model targeting one snapshot of SoC estimation is formulated. Considering temporal correlation among measurements, a detection method using the innovation test is proposed for static FDIAs, where the innovations are derived statistically. Considering the error accumulation effect, a novel sequential FDIA is proposed, which consists of a sequence of static FDIAs with small magnitudes. They can bypass most of the existing bad data detection algorithms, including the innovation test, with significant attack effects. An online approach is proposed for the practical construction of sequential FDIAs, which is formulated as a linear programming problem. Case studies based on modified IEEE 13 bus test feeder demonstrate the vulnerability of SoC estimation to FDIAs.