Deduction Simulation of Industrial Flexible Load Based on CEEMDAN-VMD-GRU

randomness and fluctuation of short-term power load are strong, and the traditional load forecasting method is difficult to grasp the law of short-term load change. In order to improve the accuracy of short-term power load forecasting, a method of short-term power load forecasting for industrial flexible loads is proposed, which Complete Ensemble Empirical Mode Decomposition with Adaptive Noise) (CEEMDAN), Varational Mode Decomposition (VMD) and Gate Recurrent Unit) (GRU). The original load sequence was extracted from the data set and decomposed into multiple intrinsic mode functions (IMF) by using CEEMDAN. Then, the components with large fluctuations were decomposed twice by VMD to effectively reduce their instability. The GRU network was used to analyze the temporal characteristics of each component, and multiple prediction results were obtained. The final load prediction results are obtained by superimposing the forecast results in reverse normalization. The proposed method is compared with the reference model and other literature methods through practical examples, and the results show that the proposed method can accurately grasp the law of load change, and the accuracy of the daily load prediction problem is 96.60%.