Load-forecasting method for IES based on LSTM and dynamic similar days with multi-features

• DOI: 10.1016/j.gloei.2023.06.003
• Publication Date: 2023-06-27T23:07:40Z
• Authors: Fan Sun; Yaojia Huo; Lei Fu; Huilan Liu; Xi Wang; Yiming Ma
• Keywords: Granularity; Feature (linguistics)

To fully exploit the rich characteristic variation laws of an integrated energy system (IES) and further improve short-term load-forecasting accuracy, a method is proposed for IES based on LSTM dynamic similar days with multi-features. Feature expansion was performed to construct comprehensive load day covering meteorological information coarse fine time granularity, far near periods. The Gaussian mixture model (GMM) used divide scene day, gray correlation analysis match granularity characteristics be forecasted. Five typical highest predicted in were selected "dynamic day" by weighting. key features adjacent forecast multi-loads using LSTM. Comparing static as input selection non-extended single features, effectiveness prediction verified.