With the deepening of the power system reform, the number of regional energy service providers will continue to increase, the grid structure will become growingly complex, and the alternative paths between power purchasers and providers will rise. In this context, a two-step selection method for regional energy service providers' trading paths is proposed. The first-step optimization selection for the trading paths is made based on a p-median model. Considering the comprehensive construction and operation cost of the energy service system, the shortest distance between an energy station and a load center, and user's load demand, a multi-objective optimization model is constructed to select sites of regional integrated energy service providers. Then, the second-step optimization selection for regional energy service providers’ trading path is made based on OD matrix and double bound gravity model. Objective functions aiming to obtain the maximum electricity transmission amounts and the maximum daily operating profits of regional energy service providers will find out the optimal energy transmission paths between regional energy service providers. The proposed models are solved by enumeration method and Dijkstra algorithm, and the calculation results are tested by a simulated case study. The simulation results show that the proposed models can effectively reduce the network loss cost, cut the abandoned wind and PV power, and improve the operating profit of regional energy service providers.