RAC-GAN-Based Scenario Generation for Newly Built Wind Farm

Due to the lack of historical output data new wind farms, there are difficulties in scheduling and planning power grid scenario generation. The randomness uncertainty meteorological factors lead results traditional generation methods not having ability accurately reflect their uncertainty. This article proposes a RAC-GAN-based method for farm output. First, Pearson coefficient is adopted this screen obtain ones that have larger impact on output; Second, based obtained factors, Grey Relation Analysis (GRA) used analyze correlation between multiple farms with sufficient (target stations), so high selected as source station. Then, K-means cluster station, thus generating target station which information serves label robust auxiliary classifier generative adversarial network (RAC-GAN) model considered basis. Finally, actual region northeast China employed arithmetic analysis verify effectiveness proposed method. It proved can effectively characteristics solve problem insufficient