At present, the network has become an indispensable part of the information society, and all kinds of institutions rely on the network for their daily work and management. At the same time, all kinds of KIS threats frequently appear, causing very serious threats. KIS technology has developed from traditional intrusion prevention and intrusion detection to intrusion tolerance and survivability research, from focusing on the confidentiality of information to focusing on the availability of information and the sustainability of services, from focusing on solving a single security problem to studying the security state and its changing trend of the whole network. Situation awareness can comprehensively grasp the current security status of the network from a macro level and overall perspective, identify potential security hazards, and speculate and estimate the future security status of the network, providing reliable reference for network managers to formulate countermeasures in a timely manner. This paper proposes a safe situation awareness prediction method based on gray clustering algorithm, and establishes a safe situation forecasting models using improved GM (1, N) theory. Through the collection of abnormal traffic in the campus network and the correlation analysis of security data, it can provide early warning and prompt for potential threats to system security. Finally, the performance of this method is compared.