基于ARIMA预测修正的工控系统态势理解算法

Situation understanding algorithm for industrial control system based on ARIMA prediction and modification

结合态势感知的概念,重点对工控系统现场控制层数据进行分析,提出一种针对工业控制系统的态势理解算法。该算法利用FCM算法实现系统正常状态空间的建模,度量出实时状态偏离正常状态的程度;此外,利用数据的时序性,通过ARIMA预测出后续时刻系统数据信息;最后使用滑动窗口技术实现对系统过去、当前和未来的数据信息融合,计算出可以表征系统实时态势的二元组,直观地呈现出系统的实时安全状况,实现当前态势理解。通过数据仿真实验,验证了算法的可执行性和有效性,该算法的输出可以为安全管理人员提供可靠的决策信息。

Based on the concept of situational awareness,this paper focused on the analysis of the field control layer data of industrial control system,and proposed an algorithm for situational understanding of industrial control system. The algorithm used FCM algorithm to model the normal state space of the system and measure the degree of real-time state deviating from the normal state;in addition,by using the time series nature of data,predicted the system data information at the subsequent time by ARIMA;finally,it used the sliding window technology to realize the data information fusion of the system in the past,present and future,calculated the binary which could represent the real-time situation of the system,and visual presentation of the realtime security situation of the system,realized the real-time situation understanding. The data simulation experiments verify the feasibility and effectiveness of the algorithm. The output of the algorithm can provide reliable decision-making information for security managers.