Wasserstein Distance-Based Renewable Scenario Generation Method
Zhi Qiang Cai, Shengnan Xu, Linghao Li, Wei Wang · 2024
In order to address the uncertainty associated with renewable energy sources, this study develops a scenario construction method. An initialization of the Gaussian Mixture Model (GMM) is achieved through the K-medoids clustering method. Subsequently, a large set of foundational scenarios generated through sampling is subjected to scenario reduction using the Wasserstein distance. The effectiveness of the proposed methodology is validated using wind and solar data from the PJM electricity market.