With the rapid growth of electric vehicle (EV) ownership, the disorderly charging behavior in residential areas has caused a serious load impact on the power grid, and it is urgent to establish a scientific and reasonable orderly charging strategy. This article proposes a dual-layer optimized orderly charging strategy and its implementation method for EVs based on a cloud-edge collaboration technology architecture. First, a hybrid prediction model of Variational Mode Decomposition Time Convolutional Network Bidirectional Gated Recurrent Unit is adopted to accurately predict wind power and achieve real-time adjustment of charging prices. Second, in the cloud-edge collaborative architecture, the cloud undertakes the scheduling function of the power grid layer, while the edge side handles the charging needs of the user layer. The optimization objective of the power grid layer is to minimize load fluctuations, while the objective of the user layer is to minimize the deviation between the actual charging power of EVs and the power guidance curve, as well as the charging cost of EVs. Under the premise of meeting the charging needs of users, the coordination and unity of individual charging behavior and power grid scheduling objectives are achieved. Finally, the parrot optimization algorithm was used to solve the problem, achieving collaborative optimization of power grid scheduling and user charging. The research results indicate that the proposed strategy can effectively reduce the peak-valley difference of the power grid load, reduce load fluctuations, and improve the capacity of new energy consumption.
Yu et al. (2026) studied this question.