With the rapid development of electric vehicles (EVs) and the expansion of the charging station network, the interaction between the transportation system and the power system is getting closer and closer. Due to the bidirectional interactions between the two systems, the small disturbances in power system may propagate to the transportation system, which may feedback to the original system to bring about cascading failures. This paper proposes a framework to analyze the cascading failures and propagation characteristics of charging stations in the coupled traffic-power system. In order to obtain the real-time traffic flow, the cell transmission model (CTM) based on dynamic traffic assignment (DTA) is employed as linear program (LP) to minimize the travel time of EVs in transportation system. Furthermore, an improved OPA (ORNL-PSerc-Alaska) model is used to simulate the cascading failures of charging stations in the power system. The proposed methods are validated using an integrated traffic and power system.