基于SVM的工控系统安全等级评估方法研究

Research on security-level evaluation method of industrial control system based on SVM

针对传统的工控系统安全等级评估方法过多依赖专家经验的不足,文章将安全等级评估问题看作是机器学习中的分类问题,提出了面向工业控制系统的安全等级评估要素体系、评估要素量化方法,并将评估要素作为分类特征在训练数据集上训练分类模型,使用分类模型实现安全等级的自动评估。实验表明,所提出的方法体系可行有效,使用支持向量机算法训练的模型在山西省工业企业工控系统安全等级评估中的分类准确率达到了90%,较好地解决了传统方法过多依赖专家的问题。

Due to the deficiency of traditional security-level evaluation methods of industrial control system relying too much on the experience of experts.This paper puts forward the evaluation factor system and evaluation factor quantitative methods for industrial control system,which regards the security-level evaluation as a classification problem in machine learning field.The classification model is trained on the training data set of taking the evaluation factors as the classification features,and the security level is automatically evaluated by using the classification model.The experiment shows that the precision of the classification model based on support vector machine reaches90%in security-level evaluation of industrial control system of ShanXi province.The deficiency of traditional methods relying too much on the experience of experts is solved by the method mentioned above.