Abstract
Recent studies imply that despite having a high level of security features and accurate algorithms, Phasor Measurement Units (PMUs) are vulnerable to False Data Injection (FDI) in the measurements. These FDIs could be: 1) an intentional FDI Attack (FDIA) or 2) unintentional errors resulting from limited precision, communication medium, noise, etc. Traditional false data detection techniques that use SCADA measurements in conjunction with the PMU measurements are insufficient to detect malicious data in PMU measurements. Thus, a new approach is proposed in this paper, which can detect the presence of false data in the PMU measurements and correct them, without relying on the use of the SCADA measurements. An affine relationship between the sparsely placed PMUs, based on the AC model of the network, is statistically extracted by offline training of a support vector regressor. This relationship is further used for the real-time detection and correction of false data in the PMU measurements. The proposed approach is tested on the IEEE 118 bus system and the Northern Regional Power Grid (NRPG) of India for detecting intentional as well as unintentional false data in single and multiple PMU measurements. Numerical results, thus obtained after testing synthetic and real-world data, are also compared with existing approaches, which justify and prove the efficacy of the proposed method.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have