Short-term Multi Load Forecasting of Integrated Energy System Based on AGCRN

Considering the dynamic coupling characteristics of multi-energy loads in integrated energy systems, a multi load forecasting model based on an Adaptive Graph Convolutional Recurrent Network (AGCRN) is proposed. Firstly, a gated recurrent unit (GRU) is utilized to mine the temporal dependencies inherent in load data. Secondly, an adaptive graph convolutional network (AGCN) is utilized to model the heterogeneity among loads and to construct a dynamic adaptive graph structure, thereby facilitating the extraction of dynamic coupling relationships among multi-energy loads. Finally, by integrating the GRU and AGCN to form the AGCRN, a multi load forecasting model is constructed. Simulation results demonstrate that the proposed model significantly outperforms traditional forecasting methods in terms of prediction accuracy.