摘要
针对工业控制网络中典型的攻击类型,提出一种利用深度学习预测工控网络通信异常的方法。首先,利用主成分分析方法对原始数据降维,消除原始数据集的相关性;其次,构建人工神经网络并利用万有引力搜索算法中粒子惯性质量计算思想改进的鱼群算法来优化极限学习机的输入权值和阈值。测试实验结果表明,异常检测的准确率有所提升,同时有效地缩短了检测时间,实现了利用深度学习预测工控网络通信异常的行为。
Aiming at typical attack types of industrial control networks, this paper proposed a method of predicting communication anomalies in industrial networks using deep learning. First, it used the principal component analysis of the raw data reduction and eliminated the correlation between the original data set. Secondly, it built artificial neural networks and to optimize the input weights and threshold limits the use of machine learning. It improved the fish swarm algorithm by the idea of particle inertia mass calculation in the gravitational search algorithm. The test experiment results show that it improves the accuracy of anomaly detection, and effectively shortens the detection time. And it realizes the purpose of making use of the depth learning to predict the abnormal behavior of communication in industrial networks.
作者
陈万志
唐雨
张静
Chen Wanzhi;Tang Yu;Zhang Jing(School of Electronic & Information Engineering, Liaoning Technical University, Huludao Liaoning 125105, China;China Petroleum Liaohe Equipment Company, Panjin Liaoning 124010, China)
出处
《计算机应用研究》
CSCD
北大核心
2019年第7期2164-2168,2178,共6页
Application Research of Computers
基金
辽宁省教育厅服务地方类项目(LJ2017FAL009)
辽宁工程技术大学博士启动基金资助项目(2015-1147)
关键词
工业控制网络
主成分分析
极限学习机
异常检测
人工鱼群算法
万有引力搜索算法
industrial control network
principal component analysis
extreme learning machine
anomaly detection
artificial fish swarm algorithm
gravitation search algorithm