Abstract: In recent years, the construction of a clean, low-carbon, safe and efficient energy system, the development of renewable energy alternatives, and the construction of a new power system with new energy as the main body have become the inevitable trend of China's energy development. In areas with abundant new energy resources, with the increasing installed capacity of new energy, large-scale comprehensive energy demonstration bases have been vigorously developed. This article establishes an evaluation system for multi-energy complementary comprehensive energy bases based on the subjective and objective weighting method, and evaluates the development and layout of multi-energy complementary bases in China's north, southwestern and eastern coastal regions. In order to further enhance the economic benefits of multi-energy complementary bases, the electricity price forecasting model based on long short-term memory neural network and the multi-energy complementary day ahead optimal dispatching model are established. The particle swarm optimization algorithm is used for optimization, so as to achieve the goal of day-ahead optimal dispatching to maximize the comprehensive income of the energy base. Finally, taking Longdong multi-energy complementary comprehensive energy base in Gansu Province as an example, the optimal dispatching simulation of typical days in summer and winter was carried out respectively. The results show that the optimized dispatching method can promote the absorption of new energy in the base and maximize the comprehensive income of the energy base, which provides support for the day-ahead optimal dispatching of large comprehensive energy bases.