Power distribution network anomaly detection method and device and nonvolatile storage medium

Abstract

The invention discloses a power distribution network anomaly detection method and device and a nonvolatile storage medium. The method comprises the following steps: collecting network traffic in a communication system corresponding to a county power distribution network; calculating traffic characteristics based on the network traffic; the traffic features are input into a target depth self-encoding Gaussian mixture model to obtain an abnormal score corresponding to the network traffic, the target depth self-encoding Gaussian mixture model is obtained by training a sample training set, and the sample training set is obtained based on an R-vine Copula model; judging whether the abnormal score is higher than a preset threshold; and when the abnormal score is higher than a preset threshold value, determining that the county power distribution network is abnormal. According to the method and the device, the technical problem that the accuracy of an anomaly detection model is relatively low due to relatively small abnormal traffic in acquired data when network attack judgment is carried out at present is solved.

Images