Source-Network-Load-Storage Cooperative Scheduling Based on Nash-Q Algorithm

With the proposal of the dual-carbon goal and the high proportion of new energy connected to the grid, in order to make full use of the source-grid-load-storage joint participation in grid flexible dispatching and new energy consumption, different forms of energy storage are set up on the three sides of the source grid load, a source-grid-load-storage collaborative optimal scheduling model aiming at low-carbon economy is constructed. Considering that the model is a multi-agent game involving new energy power stations, energy storage power stations and electric vehicles, based on Nash equilibrium theory and reinforcement learning method, this paper proposes to use Nash-Q algorithm to achieve Nash equilibrium to maximize the interests of multi-agents. Finally, using the actual regional power grid data for simulation verification, the results show that the proposed strategy can effectively reduce the economic cost and carbon emissions of the system, and improve the capacity of new energy consumption.