Medium-term forecasting of cold, electric and gas load in multi-energy system based on VAR model

A cold, electric and gas load forecasting method based on vector auto regression (VAR) in summer multi-energy flow system is presented in this paper. First, the correlation of the cold, electric, gas load and the temperature is analyzed. Then, a vector auto regression model is established, which takes the cold, electric and gas load as the endogenous variables, and the temperature as the exogenous variable to forecasting the cold, electric and gas load of the multi-energy system. By taking the load of each day as a vector and using the temperature that is closely related to the load as an exogenous variable, the forecasting model takes full account of the coupling relationship between the loads and the influence of environmental factors. In this paper, the historical data and historical temperature data of the multi-energy flow system in a certain area are used to verify the proposed comprehensive forecasting method. Compared with the single variable forecasting method, the feasibility and effectiveness of the multi-energy system load forecasting method proposed in this paper is verified.