Multiple Load Short-Term Forecasting Model of Integrated Energy System Based on the Combination of LSTM and GCN

There are not only time-domain relations but also complex spatial relations between multiple loads in integrated energy systems. Given the tight and complex relationship between multiple loads in an integrated energy system, a forecasting model based on the combination of Long Short-Term Memory (LSTM) and Graph Convolution Networks (GCN) is proposed in this paper. LSTM neural network is used to receive historical load data as input and extract temporal features, and the extracted features are used as nodes. GCN aggregates features and transmits information between loads for forecasting. The simulation results show that the proposed model has a good prediction effect.