基于改进鲸鱼算法的工控系统入侵检测研究  

Research on Intrusion Detection of Industrial Control System Based on Improved Whale Algorithm

作　　者：王华忠 程奇 WANG Huazhong;CHENG Qi(Key Laboratory of Advanced Control and Optimization for Chemical Processes of Ministry of Education,East China University of Science and Technology,Shanghai 200237,China)

机构地区：[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

出　　处：《信息网络安全》2021年第2期53-60,共8页Netinfo Security

基　　金：国家自然科学基金[61973119];中央高校基本科研业务费专项资金[222201917006]。

摘　　要：针对工控入侵检测模型训练时间长、检测率低的问题,文章提出一种改进的鲸鱼算法(IWOA)来优化SVM入侵检测模型中的参数。改进的鲸鱼算法首先引入AFSA的自适应步长和拥挤度因子,加快全局收敛速度,避免种群位置过度拥挤导致的算法早熟现象;其次,在局部搜索中加入高斯变异算子使算法跳出局部最优区域。将IWOA运用到SVM入侵检测模型参数寻优,对工控系统天然气管道数据集进行仿真,仿真结果表明,该模型检测正确率和检测速度明显提高。Aiming at the problem of long optimization time and low classification accuracy of industrial control intrusion detection model,an improved whale algorithm(IWOA)is proposed to optimize the parameters in SVM intrusion detection model.Firstly,the improved whale optimization algorithm introduces the adaptive step size and congestion factor of the AFSA.The adaptive step size can balance the ability of the whale algorithm to explore and accelerate the convergence speed.The congestion factor can avoid the premature phenomenon of the algorithm caused by overcrowding of the population search location.Secondly,the improved Gaussian mutation operator is added to the local search mechanism to make the algorithm jump out of the local optimal region.Applying it to the SVM intrusion detection model,the simulation on the natural gas pipeline data set of the industrial control system proves that the model detection accuracy rate and detection speed are significantly improved.

关 键 词：工控系统 入侵检测 鲸鱼算法 人工鱼群算法 高斯变异算子 

分 类 号：TP309[自动化与计算机技术—计算机系统结构]