Safety pre-warning method and system for industrial control network

Abstract

The invention relates to a safety pre-warning method and system for an industrial control network. The safety pre-warning method for the industrial control network comprises steps as follows: acquiring safety related data of a control event of the industrial control network; performing CEP (complex event processing) analysis on the safety related data; adopting a deep learning based classifier to perform control event risk identification on the data subjected to CEP analysis. The safety pre-warning method and system for the industrial control network can be deployed in the resource-constrained industrial control field and are applicable to a single-node or multi-node deployment environment, and the deployment way is more flexible. Furthermore, according to the method deployed in the industrial control field, network delay is reduced, processing response speed of real-time data is increased, and risk pre-warning capacity is improved. Besides, the deep learning based classifier is adopted for risk identification, accuracy rate of identification is higher, and efficiency of risk identification is substantially improved through combination of CEP analysis and deep learning scheme.

Description

technical field [0001] The present disclosure relates to the field of industrial control networks, in particular to an early warning method and system for industrial control network security. Background technique [0002] With the continuous integration of industrial control systems and the Internet, industrial control systems are inevitably exposed to network security threats, making the security and stability of industrial control networks increasingly uncontrollable. According to statistics, 70% of the networked devices in the current industrial control network have security vulnerabilities, and it is expected that 90% of the software and hardware devices connected to the Internet of Things will have a certain degree of security risks in the future. Therefore, the security issues of industrial control networks need to be resolved urgently. [0003] At present, for industrial control network security, the solutions of most manufacturers are to filter industrial control dat...

AI Technical Summary

Problems solved by technology

[0005] 1) Industrial control sites generally lack a large data center and cannot support the deployment of large cluster servers;

[0006] 2) The network structure of the industrial control site is relatively simple and single, which cannot meet the complex, large-scale, and high-speed data transmission requirements;

[0007] 3) The real-time data processing of the industrial control network generally requires a response time of about 50 milliseconds, and the above-mentioned secure big data platform has a large network delay due to the complex network structure, which cannot meet the requirements of real-time data processing;

[0008] 4) The algorithm model of the above-mentioned secure big data platform is relatively simple, and the processing capacity is limited

Images