Compound Fuzzy Clustering Anomaly Detection Based on Production Process Coupling

In recent years, the network security problem of industrial control systems is very severe. It is particularly important to detect abnormal situations in the industrial control system network correctly and timely to avoid disasters in the physical world. This paper proposes a compound fuzzy clustering algorithm, taking TE process as the research object, to analyze its working mechanism and distinguish security threats it may encounter. Considering the coupling relationship between the production process unit in the industrial control system, fuzzy clustering is used for each process unit, which performs detection and comprehensive decision-making to obtain the final anomaly detection result. Experimental results show that the proposed compound fuzzy clustering algorithm can realize anomaly detection in industrial control systems.