The regional distributed energy system is a comprehensive energy utilization system distributed on the user side, which has the characteristics of low carbon, flexibility, complementation, interconnection and so on. It is an important trend of energy system development in China in the future. Firstly, a multi-agent based information physical fusion model for regional distributed energy systems is proposed in this paper. The optimization model of the regional distributed energy system based on decision and capacity optimization is constructed at the macro and micro levels. At the same time, the regional distributed energy system is regarded as a local area “energy Internet” network. Based on graph theory and the layout optimization of an energy station, load center, energy storage center and transmission network, we constructed the “station network” layout optimization model and designed an optimization algorithm that can realize the global layout optimization of energy station, load center, energy storage center and transmission network. Finally, taking an ecological town in central China as an example, we verify the feasibility and validity of the model and method.