Font Size: a A A

Application Research Of Intrusion Detection Technology In Wind Power Generation Industrial Control Network

Posted on:2020-05-26Degree:MasterType:Thesis
Country:ChinaCandidate:D S XuFull Text:PDF
GTID:2392330623465268Subject:Computer technology
Abstract/Summary:PDF Full Text Request
Industrial control systems are an important part of the country's critical infrastructure and are at the heart of manufacturing,infrastructure,and other projects.The industrial control system is gradually integrated into the characteristics of Internet information development and sharing,forming an industrial control network.In order to meet the needs of Internet remote access,wind power industrial control networks need to provide external network related data services,which makes the security of wind farms likely to be attacked by the Internet.Therefore,this paper develops the application research of intrusion detection technology in wind power industrial control network communication data processing,and designs and implements the following related algorithms and prototype system.To solve the problem of real-time indicators in industrial control network security requirements.First,the principal component analysis method is used to reduce the dimensionality of the original data source to improve the speed of data processing.Secondly,an AMPSO-SVM algorithm is proposed based on intrusion detection technology to improve the detection effect of abnormal data in industrial control network communication data.Based on the particle swarm optimization algorithm,the adaptive mutation operation is added to effectively avoid the local optimization in the data training process,and the optimal solution obtained by AMPSO training is used as the kernel function and penalty parameter of the support vector machine.To solve the problem of data classification,an algorithm model of AMPSO-SVM-K-means++ is proposed.Firstly,the density center method is used to improve the K-means algorithm to solve the problem of the initial cluster center selection.Second,the data sorted once by the optimized support vector machine,reusing K-means++ algorithm for secondary classification to improve the detection rate of five typical data types in the classic dataset KDD Cup99.Based on the background of the actual application of the Yingren Island wind power plant in Yingkou,Liaoning Province,based on the simulation of data set,the prototype system consisting of offline training,online detection,visual display of test results and real-time working condition web services was designed and developed.The simulation results show that the paper proposed algorithm has better detection rate and false positive rate in the classical dataset KDD Cup99 than the original method.The actual test results of the prototype system show that the proposed algorithm model is applied in the wind power industrial control network,which can effectively realize the intrusion detection analysis and visual display of the on-site communication data,and provide the real-time working condition web service for Internet access.It basically meets the actual needs of the wind power industrial control network site.The paper has 32 pictures,8 tables,and 58 references.
Keywords/Search Tags:Industrial Control Network, Intrusion Detection, Principal Component Analysis, Particle Swarm Optimization, Support Vector Machine, K-means++ Algorithm
PDF Full Text Request
Related items