Guidance and regulation strategy for electric vehicle charging and discharging based on a dual-layer causal network partitioning

With the large-scale integration of EVs into distribution networks, significant progress has been made in addressing EV charging scheduling through the use of coupled power-transportation networks. However, existing methods have not taken into account the heterogeneous characteristics of the power-transportation network, and the potential impact of EV charging and discharging prices on the coupled network has not been adequately explored, limiting their effectiveness in practical applications. To enhance the coordinated optimization capability of power and transportation systems, a dual-layer causal network is proposed to achieve more precise guidance and regulation of EV charging and discharging. By constructing a power-transportation dual-layer causal network with a K-order adjacency matrix, the causal relationships within and between layers are characterized, enabling the assessment of the impact of all nodes in the dual-layer network on EV charging station load, voltage, and traffic congestion. Subsequently, power-transportation nodes that have a significant impact on charging stations are grouped into the same virtual community to determine the scope of the electricity price effects. By using the K-order adjacency matrix of a dual-layer causal network as its input, a K-GCN was designed to calculate the aggregate load and overall congestion of each virtual community, thereby improving the accuracy of the EV electricity price response model based on this information. Next, a multi-objective optimization control model is used to regulate EV loads in each virtual community, and a fairness-based compromise solution approach is proposed to balance the interests of all stakeholders. Finally, the effectiveness of the proposed control strategy is validated using existing power-transportation network topology and real datasets.