Abstract: Charging station planning (CSP) is vital to alleviating users' driving range anxiety and promoting the integration of electric vehicles (EVs). The increase in the market penetration of EVs will impact the traffic stream, leading to severe challenges in CSP. In most existing works, the traffic behavior of gasoline vehicles (GVs) and EVs, especially the interactions of route selection between the vehicles are neglected in the CSP. This paper proposes a novel traffic assignment model in which the dynamic and mixed traffic flows of GVs and EVs are taken into consideration and the characteristics of the vehicles' energy consumption are captured. Moreover, the CSP is carried out to meet the charging demands of EVs. The total cost of planning and operation of the coupling network is taken as a goal, and the growth rate of the vehicles in different planning stages, transportation and distribution operation constraints are all considered. Besides, the model is decomposed into main-sub-problems by the Benders decomposition algorithm for the global optimal solution. The simulation is carried out in the coupling network of Sioux Falls transportation network and IEEE 69-node distribution network. The results verify that the traffic flow can be improved with the dynamic mixed traffic distribution, and the charging station planning results can also be obtained optimally.