Day-ahead electricity price forecasting considering the scaled integration of renewable energy: A fusion approach based on ICEEMDAN and iTransformer
Puliang Du, Miaoheng Yang, Fei Xie, Wei Hu, Jiyu Li
Keywords:
Electricity price forecasting; Renewable energy; ICEEMDAN; iTransformer
Abstract:
As the global electricity market continues to evolve, day-ahead electricity price forecasting has become increasingly important for decision-making among various market entities. However, the continuous integration of high proportions of clean energy poses significant challenges for accurate day-ahead electricity price predictions. In response, we fully considered the coupling relationship between the output characteristics of renewable energy and the multidimensional features of electricity prices and proposed a day-ahead electricity price forecasting model based on improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) and iTransformer. First, we decompose historical electricity price data using the ICEEMDAN method to obtain multidimensional time series data based on intrinsic mode functions (IMFs). Second, we leverage the attention mechanism in iTransformer to independently predict the multidimensional time-series data containing IMFs and renewable energy output, forming a forecasting model suited for the large-scale integration of renewable energy in the electricity market. Finally, using historical electricity price and renewable energy output data from Spain as a case study, we constructed a simulation model. The results demonstrate that the ICEEMDAN-iTransformer model effectively handles noise, nonlinearity, and non-smoothness in data following the integration of renewable energy, enabling more stable and accurate forecasting results.
Fulltext download:
 Day-ahead electricity price forecasting considering the scaled integration of renewable energy: A fusion approach based on ICEEMDAN and iTransformer [PDF file] [Filesize: 1.7 MB]
 10.7441/joc.2025.01.10
Du, P., Yang, M., Xie, F., Hu, W., & Li, J. (2025). Day-ahead electricity price forecasting considering the scaled integration of renewable energy: A fusion approach based on ICEEMDAN and iTransformer. Journal of Competitiveness, 17(1), 201-232. https://doi.org/10.7441/joc.2025.01.10
|
