Industrial control systems are an essential component of national infrastructure, widely used in industries such as energy, manufacturing, transportation and defense. They represent a critical resource that directly impacts the country's economy and people's livelihoods. Industrial control protocols serve as vital links, enabling real-time data exchange, data acquisition, parameter configuration, status monitoring, anomaly diagnosis, command issuance, and execution in control systems. The security of these protocols is closely related to the reliable and stable operation of industrial control systems. Detecting anomalies in industrial control networks is a crucial aspect of ensuring the stable operation and security of industrial automation systems. This paper focuses on researching methods for detecting anomalies in industrial control networks, with a primary emphasis on intrusion detection. We construct an industrial control network device fingerprint library based on behavioral patterns and combine ISSA and TWSVM algorithms to conduct research on intrusion detection methods in industrial control networks. The aim is to achieve functions such as secure monitoring, anomaly alerting, and threat warning for industrial control networks, and promote the application of this system in various industrial control industries.