Research On Security Situation Awareness Methods For Industrial Internet Based On Machine Learning

As an important part of the new generation of information infrastructure,Industrial Internet is facing ever-increasing security issues while promoting Industrial digitization,intelligence,and networking.Security situation awareness technology is helpful to grasp the current situation of Industrial Internet security,predict the future development trend of security,and help security decision makers to take appropriate protective measures,which is an important method to establish an active defense mechanism for Industrial Internet security.This thesis studies the related methods of network security situation awareness technology,proposes the security situation assessment and situation prediction method of Industrial Internet based on machine learning and establishes the security situation awareness model for Industrial Internet.The main research contents are as follows:1.In the process of Industrial Internet security situation assessment,due to the wide variety and complex composition of security data,it is difficult to assess the security status.This thesis proposes an Industrial Internet security situation assessment method based on the combination of principal component analysis and least square support vector machine algorithm.Firstly,the method uses the principal component analysis to reduce the safety samples and extract the principal components.Secondly,use the improved particle swarm algorithm to optimize the parameters of the least square support vector machine,and perform classification training on the obtained principal component data.Then,the Industrial Internet security situation value is calculated through the established security situation assessment model and the security situation quantification method.Finally,the simulation experiment shows that the evaluation accuracy of the method in this paper has reached 97.22%.2.Aiming at the problem of low accuracy of security situation prediction,an improved grey wolf algorithm is proposed to optimize support vector regression for Industrial Internet security situation prediction.Firstly,this thesis proposes a new calculation formula of nonlinear convergence factor and synergy factor to improve the iterative process of grey wolf algorithm,and uses the differential algorithm to solve the premature problem of grey wolf algorithm.Secondly,the improved gray wolf algorithm is used to optimize the relevant parameters of the support vector regression prediction algorithm to establish the optimal security situation prediction model.Then,the optimized support vector regression model is applied to the Industrial Internet security situation prediction.Finally,a comparative experiment is carried out with the data of security situation value.The simulation results show that the prediction accuracy of the proposed method has reached 91.67%.In short,the security situation assessment method and the security situation prediction method based on machine learning can ensure the effective assessment of the current security situation in Industrial Internet and predict the trend of the security situation in the future while improving the accuracy rate.