Abstract: Aiming at the uncertainty of the time distribution of charging demand in vehicle-to-grid (V2G) scheduling and the uncertainty of output under the penetration of high proportion of new energy, which makes it difficult to balance the load of the grid, a V2G collaborative optimization strategy driven by electricity price response under a high proportion of new energy access is proposed. Firstly, a V2G collaborative optimization model driven by electricity price response for high proportion new energy access scenarios is constructed, which guides the charging and discharging behavior of V2G charging piles and electric vehicles by regulating electricity prices, and is optimized with multiple goals to minimize the total cost of the power grid, minimize the peak-to-valley difference of grid load, and maximize user benefits. Secondly, the non-dominated sorting genetic algorithm Ⅲ(NSGA-Ⅲ) algorithm is fused with the Wasserstein generative adversarial network(WGAN) to establish a dynamic adjustment algorithm with non-dominated sorting and crowding distance selection elite solution reward function for solving the model. Case analysis shows that compared with the traditional NSGA-Ⅲ algorithm, the proposed NSGA-Ⅲ WGAN algorithm can effectively reduce the operating cost of the power grid, improve the new energy consumption capacity of the power grid, stabilize the peak-to-valley difference of the load of the power grid, and improve the benefits of users through efficient global search and local fine-tuning capabilities.

Key words: V2G charging piles, node electricity price regulation, NSGA-Ⅲ algorithm, WGAN algorithm