Industrial control safety detection method based on feature expansion

Abstract

The invention discloses an industrial control security detection method based on feature expansion, and the method comprises the steps: S1, selecting a plurality of pieces of feature data from network connection data, calculating the average time T of the first n pieces of data of a certain piece of feature data at the time t0, n being greater than or equal to 4, and constructing a time feature = data time t0 -average time T; using the constructed time features to construct a data matrix, and enabling the constructed data matrix with the time features ito be as a group of data; then, according to the above operation, constructing a second group of data for the selected next piece of feature data until the plurality of pieces of selected feature data are constructed to obtain a plurality of groups of data, and completing a first preprocessing step; and s2, inputting a plurality of groups of data obtained after the first preprocessing into a model for training, wherein the trained data is used for executing an industrial control safety detection task and the like. According to the method, feature selection is simplified, feature enrichment is formed, multi-dimensional pre-judgment and recognition are achieved, the detection precision is improved, and the training speed is increased.

Description

technical field [0001] The present invention relates to the field of industrial information security, and more specifically, relates to an industrial control security detection method based on feature expansion. Background technique [0002] Hacking attacks on industrial control systems are showing an increasing trend due to the informatization and networking of industrial production. The characteristics of industrial control systems, such as production continuity, equipment complexity, industry process differences, strong coupling of product components, network protocol diversity, manufacturing technology monopoly, and indifference of personnel security information awareness, lead to difficulties in the protection of industrial control system information security. . [0003] The existing artificial intelligence-based industrial control network data detection is mainly divided into traditional machine learning detection and RNN related and variant detection based on natural...

AI Technical Summary

Problems solved by technology

Most of its preprocessing modules are not feature selected, and its feature selection is selected for professional knowledge. Therefore, the preprocessing module will cause multi-dimensional useless features to affect other features, and also increase the computational complexity of feature training.

Images