Multi‐Objective Optimization Study of Time‐of‐Use Electricity Price for Electric Vehicles Based on WGAN‐GP

ABSTRACT The growing penetration of variable wind and solar generation poses operational challenges for power grids, while the rapid adoption of electric vehicles (EV) further intensifies system load stress. This study develops an optimized time‐of‐use (TOU) pricing strategy that incentivizes EV owners to shift charging to off‐peak periods and curtail charging during peak demand, jointly minimizing system load volatility and user charging costs. A Wasserstein GAN with gradient penalty (WGAN‐GP), an enhancement over the original WGAN, is employed to synthesize high‐fidelity wind–solar generation scenarios that serve as reliable inputs for tariff optimization. A moving boundary method is applied to segment EV charging demand into dynamic time periods. Building on these components, we formulate a bi‐objective TOU pricing model that explicitly incorporates EV users' price responsiveness. Case studies demonstrate the superior scenario generation performance of WGAN‐GP and the tangible benefits of the proposed TOU strategy. The optimized pricing achieves a 9.09% reduction in average user charging cost and a 24.11% reduction in the peak–valley load gap, thereby mitigating system stress, enhancing the operational flexibility of the virtual power plant (VPP), improving renewable (wind–solar) utilization, reducing reliance on energy storage systems (ESS), and increasing the wind–solar absorption ratio.