Short-Term Load Forecasting Based on Ceemdan and Transformer

Short-term load forecasting (STLF) is an essential part of energy plan, and it is very meaningful for energy management. Recently, some deep learning models have been prominent in load forecasting. This study focuses on the Transformer model which can solve the long memory loss problem by introducing the attention mechanism. A hybrid model named CEEMDAN-SE-Transformer incorporating Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN), sample entropy (SE) and Transformer is proposed. To demonstrate the superiority of the proposed CEEMD-SE-Transformer model, a variety of machine learning models are used as a comparison, the one with the best performance is used to compare with the Transformer model and the proposed CEEMD-SE-Transformer model. At the same time, the Empirical Mode Decomposition (EMD) technique is used to compare with CEEMDAN and validate the superiority of CEEMDAN-SE-Transformer. In this study, short-term load forecasting (1h, 4h, 8h, 12h, and 24h, respectively) is performed for New York City. The results show that the CEEMDAN-SE-Transformer gets the best forecasting results of all the comparative models, taking the prediction result of 24h as an example, the mean absolute percentage error (MAPE) is 4.83%, root mean square error (RMSE) is 1.27 and mean absolute error (MAE) is 0.96.