A new method of short-term load forecasting based on the selection of compound variables

Abstract

Accurate short-term load forecasting is an important guarantee for achieving lean operation and management of the power grid. However, there are difficulties in precise forecasting such as short-term load variability and selecting key factors of load forecasting. We establish a short-term load forecasting method based on variational modal decomposition and compound variable selection. The method is used to decompose the original power load data into multiple sub-sequences, and mine short-term load change characteristics while avoiding mode aliasing problems. Meanwhile, a complex variable selection algorithm is proposed to analyze and screen the key factors affecting load changes, effectively eliminating undesired data and further simplifying the complexity of the prediction model. Each sub-sequence is predicted and merged through the long and short-term memory neural network, achieving the final short-term load forecast. The verification results of the actual data of Changsha City selected for the entire year of 2019 show that the algorithm proposed here can accurately select the key influencing factors of load forecasting under complex external influence factors. With traditional forecasting models, the proposed model structure is simpler. The prediction accuracy is higher.