pone.0338321.t004 -

Under the imperative of achieving dual-carbon goals, the number of distributed energy resources are gradually increasing, thereby amplifying the challenges to grid stability and power balance. Consequently, there is an urgent need to leverage the potential of source-network-load-storage for enhanced power regulation and control. This paper proposes a cloud-edge-end-based multi-time scale economical management of virtual power plant (VPP) source-network-load-storage in the context of electricity-carbon market. In the first layer, a cloud-edge scheduling approach is used to optimize the source-network-load-storage system of the VPP over a long time horizon, aiming to maximize economic benefits. In the second layer, a novel real-time pricing mechanism is employed to effectively manage and regulate the electric vehicle (EV) storage and charging stations. After obtaining the economic management parameters from the previous layer, the second layer employs a real-time scheduling approach based on end-side model predictive control (MPC), to address multi-energy supply-demand fluctuations. To achieve efficient solution, the original two-layer optimization problem is reformulated using mixed-integer linear programming (MILP). Comparative analyses have demonstrated the superior economic and practical performance of the proposed two-layer optimization approach. Simulation results indicate that the total operating cost of the system can be reduced by 2.35%, with a higher flexibility of electricity market operations.