Industrial equipment abnormal behavior detection method based on equipment behavior recognition graph

Abstract

The invention relates to an industrial equipment abnormal behavior detection method based on an equipment behavior recognition graph. The method comprises the following steps: constructing an equipment behavior recognition graph at a current moment, wherein nodes in the equipment behavior recognition graph are attributes of behavior events, edges in the equipment behavior recognition graph are co-occurrence relations between any two attributes, and each edge has a weight; introducing an initial weight mechanism, a weight smoothing mechanism, a time influence mechanism and a period influence mechanism, and carrying out weight transformation on the equipment behavior recognition graph; capturing equipment behavior event representation of the equipment behavior recognition graph after the weight transformation through an event representation extractor; and taking the equipment behavior event representation as input of a trained multi-layer perceptron to obtain a probability that current industrial equipment has an abnormal behavior. Compared with the prior art, the method has the advantages of improving the accuracy of intercepting abnormal behaviors and the robustness of a model and the like.

Description

technical field [0001] The invention relates to the technical field of industrial Internet abnormal behavior detection, in particular to a method for detecting abnormal behavior of industrial equipment based on a device behavior recognition graph. Background technique [0002] With the increasing expansion of Internet business, the industrial scene has also entered a new stage of development. The Internet connects the industrial control system and the cloud platform as a whole, realizing the comprehensive perception of industrial data. The complexity of the Internet environment increases the possibility of industrial control systems being attacked by the outside world. Abnormal behavior of industrial equipment often brings uncontrollable huge losses to enterprises. In order to ensure the business security of industrial control systems in industrial Internet scenarios, it is necessary to establish Practical and effective detection method for abnormal behavior of industrial eq...

AI Technical Summary

Problems solved by technology

However, the current scale of industrial Internet data is often large, and abnormal behavior data is usually hidden in massive normal data, which poses a huge challenge to the detection of industrial Internet abnormal behavior

Due to the deep hiding of abnormal data, it is difficult to mine abnormal features, which in turn affects the performance of machine learning models

Images