Network Security Situation Assessment And Prediction Of Industrial Internet Based On Belief Rule Base

Industrial Internet is a global network formed by the increasingly close combination of industry and Internet under the continuous development of information technology.With the emergence of new technologies and development concepts,industrial systems are transforming from traditional information technology applications to more advanced digitization,networking and intelligence,but this also brings many security risks to the Industrial Internet.The security situation awareness technology can evaluate the security status of the Industrial Internet in a timely manner,which can predict the development trend of the future security status,and can also improve the risk monitoring capabilities,emergency response capabilities and active defense capabilities of the Industrial Internet.Therefore,this paper proposes corresponding assessment and prediction methods for the network security situation of the Industrial Internet,and establishes an assessment and prediction model.The specific content is as follows:Aiming at the problem of Industrial Internet security situation assessment,this paper adopts the evidence reasoning（ER）algorithm and the belief rule base（BRB）method to establish the Industrial Internet security situation assessment model.Firstly,the influencing factors of Industrial Internet security situation are analyzed,and evaluation indicators（including quantitative data and qualitative knowledge）are selected.Secondly,combined with expert knowledge,the ER algorithm is used to fuse the evaluation indicators,and an Industrial Internet security situation evaluation model based on ER and BRB is established according to the fusion results.The projection covariance matrix adaptive evolution strategy（P-CMA-ES）is adopted to optimize the model parameters.Finally,the validity and superiority of the method are verified through the case analysis of Industrial Internet security situation assessment.Aiming at the problem of Industrial Internet security situation prediction,this paper adopts the prediction method based on time series analysis（TSA）and BRB,and the Industrial Internet security situation prediction model based on TSA-BRB is established.Firstly,a BRB-based Industrial Internet security situation prediction model is established by combining expert knowledge and quantitative data processed by TSA sliding window mechanism.Secondly,nonlinear optimization algorithm（NOA）is used to optimize the interval reference provided by experts,and P-CMA-ES is used to optimize other parameters of the model.Finally,the validity and superiority of the method are verified through the case analysis of Industrial Internet security situation prediction.The research results show that the above methods have strong applicability to the assessment and prediction of Industrial Internet security situation,which can accurately reflect the security situation of Industrial Internet.The MSE values of the assessment and prediction results both reach 10^-3magnitude,which is one magnitude higher than that of other models.This can provide safe and reliable suggestions for network administrators,make adequate preparations for future insecure incidents in advance and take correct countermeasures in time,so as to improve the risk monitoring capability,emergency response capability and active defense capability of the industrial Internet.