文献类型：期刊文献

中文题名：融合PCA和PSO-SVM方法在工控入侵检测中的应用

英文题名：Application of Fusion PCA and PSO-SVM Method in Industrial Control Intrusion Detection

作者：王华忠[1];杨智慧[1];颜秉勇[1];陈冬青[2]

机构：[1]华东理工大学化工过程先进控制和优化技术教育部重点实验室,上海200237;[2]中国信息安全测评中心,北京100085

年份：2017

卷号：33

期号：1

起止页码：80

中文期刊名：科技通报

外文期刊名：Bulletin of Science and Technology

收录：CSTPCD;;北大核心:【北大核心2014】；

基金：国家自然科学青年基金(51407078)

语种：中文

中文关键词：工业控制系统;入侵检测;PCA;粒子群算法;支持向量机

外文关键词：industrial control system;anomaly detection;PCA;particle swarm optimization;support vector machine

摘要：有效防御病毒对工控系统的入侵是目前工控安全研究的难点问题。为了提高工控系统入侵检测的准确率,本文设计提出了一种主成分分析(PCA)与PSO-SVM相结合的工控入侵检测方法。针对工业控制系统网络数据高维的特性,该方法利用PCA对采集的网络入侵数据进行数据降维与特征提取,支持向量机(SVM)入侵检测的性能主要取决于核函数参数取值的优劣,采用粒子群算法(PSO)对支持向量机参数进行优化,以获得最优的SVM工业控制系统入侵检测模型。采用密西西比州立大学关键基础设施保护中心最新提出的工控标准数据集进行仿真实验,结果表明该算法在攻击检测与攻击类型识别方面均有较高的查准率,提高了工业控制系统的安全性能。
The effective defense against virus invasion of industrial control system is a difficult problem in the research of industrial control security. In order to improve the accuracy of intrusion detection in industrial control system, this paper presents a method of industrial intrusion detection based on the combination of principal component analysis (PCA) and PSO-SVM. For industrial control network data are high dimensional characteristic, the method using PCA on the acquisition of the network intrusion data to reduce data dimension and feature extraction, support vector machine (SVM) intrusion detection performance depends on the kernel function parameters of the pros and cons, using particle swarm optimization (PSO) algorithm is used to optimize the parameters of support vector machine, in order to obtain the optimal SVM industrial control system intrusion detection model. The Mississippi State University Center for critical infrastructure protection the latest proposed industrial standard data sets to carry on the simulation experiment. The results show that the algorithm in attack detection and attack type recognition has a high precision and improve the safety performance of the industrial control system.