Objective In the context of the "dual carbon" goals, a high proportion of photovoltaic systems and electric vehicles can significantly impact the power grid. Building clean, efficient, and stable photovoltaic storage charging stations is essential for achieving green and low-carbon development. To address the optimization of photovoltaic storage charging station configurations, a two-layer optimization method based on orderly EV charging has been proposed.

Method This method involves constructing a typical photovoltaic output scenario library using generative adversarial networks and an improved K-means++ algorithm, and predicting EV charging loads with a LSTM-Transformer hybrid model. Time-of-use pricing mechanisms are used to guide orderly charging to optimize the charging load curve. Secondly, a dual-layer optimization model for the planning and operation of photovoltaic storage charging stations is established. The outer layer aims to minimize the total lifecycle cost through capacity configuration, while the inner layer formulates operational strategies based on economic efficiency and reliability. This dynamic coupling between planning and operation is achieved through bidirectional parameter exchange. Finally, an improved grey wolf optimization algorithm is used to solve the model, resulting in the optimal configuration of photovoltaic storage charging stations.

Result Case studies show that the proposed configuration strategy achieves optimal economic benefits, effectively improves the voltage quality and stability of the power grid, and enhances the operational level of the distribution network.

Conclusion The research results can provide certain reference for improving the operation level of the distribution network.