In the background of "carbon peak, carbon neutral", large-scale new energy grid connection has caused the system peak adjustment pressure, new energy consumption difficulties and other problems. With the increasing proportion of flexible load, a two-layer optimal operation model of "source-network-charge-storage-coordinated" was proposed to solve the problems faced by the system according to the different characteristics of flexible load. Firstly, based on the time-of-use tariff strategy, some users were guided to use electricity, and the upper multi-objective model was established to improve the comprehensive satisfaction of both sources and loads and to make fluctuation of net load minimized. Particle swarm optimization algorithm was used to solve the upper level model and to obtain the optimized load demand response in order to bring it into the lower level model. Secondly, the operation mode of flexible load participating in scheduling was determined and the lower layer model aiming at the minimum operation cost of the system scheduling was constructed in the light of the supply and demand relationship between the two sides of the source and load. In addition, the Cplex solver was used to solve the model. Finally, taking the actual system of a region as an example, the results show that the participation of flexible load in power grid dispatching can alleviate the peak regulation pressure of power grid, improve the absorption capacity of renewable energy, and also verify the effectiveness of the model proposed.