An unsupervised cyberattack detection scheme for AC microgrids using Gaussian process regression and one‐class support vector machine anomaly detection

This paper addresses the cybersecurity of hierarchical control AC microgrids with distributed secondary control. The false data injection (FDI) cyberattack is assumed to alter operating frequency inverter-based generators (DGs) in an islanded microgrid. For consisting grid-forming inverters a manner, attack on one DG deteriorates not only corresponding but also other DGs that receive corrupted information via communication network. To this end, FDI detection algorithm based combination Gaussian process regression and one-class support vector machine (OC-SVM) anomaly introduced. unsupervised sense it does require labelled abnormal for training which difficult collect. model predicts response DG, its prediction error estimated variances provide input OC-SVM detector. returns enhanced performance than standalone OC-SVM. proposed detector trained tested collected from 4 microgrid test validated both simulation hardware-in-the-loop testbeds.