Abstract: With the increasing number of electric vehicles,the planning of electric vehicle charging station（EVCS）is extremely urgent. Due to the randomness of traffic flow and vehicle owner’s travel situation,the EVCS charging load is also uncertain,so the electrical connecting nodes and capacity of the EVCS in power grid will affect the economy and safety of the distribution network. Based on the sequential feature and correlation of the EVCS charging load and several typical distributed energy resources,a large number of scenarios are generated by Latin hypercube sampling（LHS）,and reduced scenarios are obtained by K-means clustering algorithm which introduces a margin ranking factor. In addition,an optimization model is established by using the comprehensive evaluation index composed of the minimum active power loss and voltage deviation of the distribution network as an objective function. Finally,IEEE 33-bus system is used to analyze the examples,obtaining the optimal electrical connecting nodes and capacity of the EVCS.