Abstract: The increase in the number of electric vehicles (EVs) and the enhancement of their endurance make the coupling between EV, transportation networks, and power grid more complex. Therefore, how to accurately describe the charging demand of EVs under complex coupling conditions and balance the interests of charging station operators and EV users is a problem that must be considered in the planning of EV charging stations. Therefore, the Monte Carlo method is used to obtain the charging demand of each EV in the planning area under typical scenarios, and the charging power of different road nodes in each period is clustered to the corresponding cluster center node, then its probability density function is obtained by using the Gaussian mixture model. A bilevel optimization model for EV charging station planning that comprehensively considers the interests of charging stations and users is established. Based on complex network theory and voltage sensitivity index, candidate charging station nodes are selected from the perspectives of transportation network and power grid, and the optimization problem is solved by slime optimization algorithm. A coupled network composed of a 245-node road network and an IEEE30-bus power grid is taken as an example, and the comparison results verify that the proposed planning method can be adopted to not only retain the spatiotemporal distribution characteristics of EV charging demand, but also facilitate a win-win situation for charging stations and users.