Smart FDI Attack Design and Detection with Data Transmutation Framework for Smart Grids

Communication Network Layer State Estimation Measurement Model for a Cyber-Secure Smart Grid

“…Otherwise, this sample is classified to be normal, and the mean, µ m and inverse covariance matrix, Σ −1 m are updated using the Woodbury Matrix Identity [15]. The CECD-AS uses an adaptive threshold, which is updated in an online sliding window fashion [10], unlike the fixed threshold in the CorrDet detector.…”

“…The combined distance measure obtained from the hybrid model is compared with a threshold value, χ 2 (d−N ),p , which is based on the confidence level as expressed in (15), where (d − N ) denotes the degrees of freedom and p is the probability with a value of 0.97, similar to [16]. A given sample is classified to be abnormal if the threshold value is exceeded.…”