An information security risk assessment method for an industrial control system

Abstract

The invention relates to an information security risk assessment method of an industrial control system, comprising the following steps: S1, obtaining a risk assessment value and a plurality of sets of assessment scores corresponding to a plurality of risk assessment elements as an initial sample data set; S2, the initial sample data is processed by KPCA, and the principal component is selected according to the cumulative contribution rate to obtain the reduced-dimension sample data set; S3, training the BP neural network optimized by the genetic algorithm to obtain a prediction model by taking the reduced-dimension sample data set as a training sample; S4, inputting evaluation scores of a plurality of risk evaluation elements into a prediction model to obtain a prediction value of a riskevaluation value. Compared with the prior art, the invention not only improves the problem of parameter selection in the neural network, but also effectively improves the evaluation precision of the risk evaluation model, which is the combination of conventional modeling and intelligent method, and has important significance for the industrial control system.

Description

technical field [0001] The invention relates to the technical field of information security, in particular to an information security risk assessment method for an industrial control system. Background technique [0002] Industrial Control System (ICS for short) is widely used in petrochemical, transportation, water treatment and other national key infrastructures. With the development of information technology and the deepening of the integration of "industrialization and industrialization", traditional industrial control systems, IT systems, and even the Internet are increasingly connected, resulting in increasing security threats to ICS. [0003] In recent years, the overall scale of the industrial control system is gradually expanding, and the security level related to it has also been comprehensively improved. However, it should not be overlooked that the industrial control system under the new situation is still difficult to eliminate information risks from the source...

AI Technical Summary

Problems solved by technology

BP neural network is the most widely used prediction model, but this model has two obvious shortcomings: one is easy to fall into local minimum; the other is slow convergence

Images