Abstract: It is one of the key tasks of current railway network security construction to study the application of network security situation awareness technology based on big data in railway so as to endow the railway information network with the ability of comprehensive awareness and proactive early warning. Therefore, it was aimed to solve the lack of timely surveillance and awareness over potential threats that might impose limitations on the development of application systems. Meanwhile, the problems of high false alarm rate and missing alarms due to huge amount of data, complexity of business, hierarchical overlapping existing in traditional situation awareness platform were also dealt with. Based on high-speed access of big data, artificial intelligence and parallel processing technology were used to optimize the perception and prediction algorithm and a solution of network security situation awareness platform suitable for Chinese railways was also put forward. Furthermore, the solution had been tested and verified in a test environment for a month and the results show that this solution is superior to traditional network security situation awareness platform in perceiving and discovering potential security threats in a railway information network environment with the characteristics of high throughput and high complexity, and can effectively reduce the rate of false alarm and improve the quality and level of alarming of the Chinese railway information network with the characteristics of high throughput and rapid real-time response.