Anomaly Detection for Industrial Control Systems Using K-Means and Convolutional Autoencoder

Abstract

With the rise of Internet of Things technology and smart manufacturing, the issues of cyber security in industrial control systems (ICS) have become of great concern. Critical infrastructures, such as power plants and water treatment facilities, which heavily rely on ICS, have increasingly come under cyberattack in recent years. These cyberattacks pose severe threats to public safety as well as national security. Therefore, protecting ICS from cyberattack is an important research issue in the era of information technology and operation technology convergence. In this study, we propose an anomaly detection method based on semi-supervised techniques for industrial control systems, using the k-means and convolutional autoencoder methods. Experimental results show that the proposed method outperforms the other methods in almost all performance metrics for gas pipeline and water storage tank datasets.