Short-term load forecasting based on sample weights assignment

• DOI: 10.1016/j.egyr.2022.09.101
• Publication Date: 2022-10-20T11:42:09Z
• Authors: Mingxu Xiang; Sijun Du; Juan Yu; Yi Xu
• Keywords: Sample (material); Similarity (geometry)

Short-term load forecasting (STLF) is the basis of power system operation. Considering that importance different training samples different, a sample weights assignment method proposed in this paper to help STLF learn key sample. At first, similarity measured considering characteristics input components. Based on this, are selected. Finally, assigned with through designed function. With method, model able focus crucial samples. Simulation results data-driven models demonstrate effectiveness method.