Survey on Methodology of Intrusion Detection in Industrial Control System Based on Artificial Intelligence

Abstract

With the increasingly close connection between the industrial control system and the Internet, the industrial control system are facing more and more cyber threats. Intrusion detection technology is one of the important technologies to protect the security of industrial control systems. With the development of artificial intelligence, many researchers have applied artificial intelligence algorithms based on machine learning and deep learning in the industrial control intrusion detection systems. Compared with the traditional methods, the detection accuracy based on machine learning is higher, but the procedure of manual feature extraction limits its practical application. The detection method based on deep learning has a strong feature extraction ability, but its computational complexity is high. In many cases, the detection effect is not ideal in industrial control scenes with real-time requirements. Based on the analysis of the related research, this paper not only summarizes the current research status of intrusion detection in industrial control system based on artificial intelligence, but also deeply discusses the challenges faced in this field. It is pointed out that the intrusion detection algorithm for the industrial control system should be further improved according to the characteristics of the industrial control system, such as real-time requirement, limited computing resources, high data dimension, and high data noise.