Short-Term Load Forecasting Method Based on VMD and Rank Set Pair Analysis

The short-term load forecasting of power system is crucial for the safety, economy, stability, and reliability of power system operations. In this regard, this paper proposes a short-term load forecasting model based on variational mode decomposition (VMD) and rank set pair analysis (R-SPA). Initially, the load original sequence is decomposed into low-frequency components containing the main features, mid-frequency components containing detailed features and random components containing stochastic features of the original load by using VMD. Subsequently, the R-SP A model with different set capacities is employed to predict each component, followed by the reconstruction to obtain the final prediction value. Finally, the short-term load forecasting is accomplished through a rolling prediction approach. Simulation results demonstrate that the proposed forecasting model has superior predictive performance.