High-speed rail (HSR) has become a critical infrastructure and even has formed a network in some countries, which plays an increasingly significant part in the modern transport system and exerts increasingly obvious impacts on regional economic activity and human mobility. Although its interruption may cause serious consequences, researching its robustness doesn’t attract attention until recent years. Based on previous researches on the issue, this paper proposes a comprehensive method to assess the robustness of high-speed rail network (HSRN) by combining the network function features with network topology features and modeling HSRN as a two-layer network based on the high-speed trains running in the network, in which the travel time and passenger flow are taken into account. Finally, taking China’s HSRN (CHSRN) as a case study, this paper investigates the failure effects of HSR stations and high-speed trains on the robustness of CHSRN, especially the space-time failure effects of high-speed trains at three typical HSR stations. Results reveal that the failure effects of trains at the selected stations on the performance of CHSRN are limited, and the failure effects of trains at three typical stations on the performance are different as the different failure time, which is beneficial to prioritize the decisions on maintenance resource allocation and prepare emergency response plans.