Guaranteed Data Recovery and Oscillatory Mode Estimation in Power System Using Novel RPCA

Abstract

Presence of malicious injection occurs as the con-sequence of cyberattacks or PMU malfunctioning and due to which the signals obtained from PMUs may have an impact on the stability & functionality of the power networks. A technique for detecting phony data in PMU measurements is presented in this paper. A robust Principal Component Analysis (RPCA) with Probabilistic Distributional Clustering (PDC) based technique is used to recover the malicious signals in the compressed sensing problem formulation for bad data detection. Lastly, esprit based method is utilized for mode estimation. By doing a statistical analysis research on a synthetic signal using Monte Carlo simulation, the efficacy & resilience of the Novel Robust PCA technique are confirmed and contrasted with other recently discovered methods. Additionally, real probing data and two area data from the Western Electricity Co-ordinating Council (WECC) are used to validate this technique.