To ensure the security of the coupled distribution-transportation networks (CDTN) after a fault occurs, an improved road model and a bi-level planning method are proposed for electric vehicles charging stations (EVCSs). This approach considers the feature of fault propagation in the CDTN. Specifically, the upper-level model identifies the optimal locations and corresponding capacity strategies. In the lower-level model, EV charging routes are designed to reduce the overall journey costs for electric vehicles (EVs). The outputs from the lower-level model, including traffic flow and EV queuing results, are incorporated into the upper-level model for refinement. The proposed bi-level model is iteratively solved, with the optimal sensitivity coefficient serving as the evaluation criterion. Finally, this paper utilizes a CDTN, consisting of a 90-node regional road network and three IEEE 33-node systems, to verify the accuracy and feasibility of the proposed model. The optimal planning results for charging stations within the example area indicate the allocation of 10 stations, with a final total investment of 21.2721 million CNY, which is significantly better than other planning schemes. Moreover, all schemes are verified through N-1 fault simulation to determine the capacity allocation for each charging station.