基于PSO寻优算法的过程控制网络异常检测技术

Based on PSO Optimization Algorithm Process Control Network in Anomaly Detection

针对目前工业过程控制网络通信异常检测多数是人工为主,效率较低的问题,提出一种基于PSO优化算法,设计了一种自动化的过程控制网络监视和异常诊断系统;该方法通过深入研究过程控制网络的工作原理,总结过程控制网络中故障的特征,把过程控制网络中的故障信息引入优化的POS决策树,通过计算故障特征的信息熵作为核心的数据,进行异常监测挖掘,总结学习网络故障的规律,得出检测过程控制服务器和工业设备之间通信故障早期预兆的报警条件;结果表明,这种方法能够提高大型过程控制网络的异常监测准确性7个百分点。挖掘时间也缩短了13%。

In view of the present industrial process control network communication anomaly detection majority is artificial is given priority to, low efficiency, based on the PSO algorithm is proposed in this paper, design a kind of automated process control network monitoring and abnormal diagnosis system. This method through indepth study process control network of the working principle, summarizes process con trol network fault characteristics, the process control network of the fault information and to optimize the introduction of POS decision tree, through the calculation of the failure characteristics of information entropy as the core data, conduct abnormal monitoring mining, summari zes the law of learning network fault, it is concluded that detection process control server and industrial equipment of communication between fault early warning alarm condition. The results show that this method can improve the large process control network anomaly monitoring ac curacy of 7%. Mining time also was reduced by 13%.