Renewable Energy Hosting Capacity Evaluation of Distribution Network Based on WGAN Scenario Generation

With the continuous improvement of the proportion of renewable energy access in the distribution system, how to reasonably consider the uncertainty of renewable energy power output and efficiently and accurately evaluate the renewable energy hosting capacity of the distribution network is an urgent issue to be considered in current power grid planning. In this work, we proposed a scenario generation method based on Wasserstein Generative Adversarial Networks for renewable energy power output, which can generate the scenarios with the same distribution as real data and more diversity. Then, we extracted representative typical scenarios of renewable energy based on the improved K-means algorithm. We also established a mathematical model for calculating the hosting capacity of the distribution network with the maximum installed capacity of renewable energy as the objective function, including constraints such as power flow balance, voltage deviation, and voltage fluctuation. Finally, based on wind and solar power generation data for one year from NREL laboratory , a simulation calculation was conducted for a 45-bus distribution network to verify the effectiveness of the proposed model and method.