Forecasting Heating and Cooling Energy Consumption by Seasonal ARIMA Models

The main purpose of this paper is to predict energy consumption of heating/cooling for the building’s sector based on Seasonal Auto Regressive Integrated Moving Average SARIMA model for a tertial building located in Errachidia city in Morocco. The data of heating and cooling consumed energy have been collected during the last five years from January 2016 to December 2020 by using ECOTECT ANALYSIS software, considering the climate conditions plus the occupants’ behaviors. After fitting and testing SARIMA model, the analysis of the Mean Absolute Error MAE, Mean Square Error MSE, Root Mean Square Error RMSE and the Scatter Index (SI) show that the seasonal ARIMA approach gives optimal results for both cooling and heating energy consumption forecasting. Finally, this developed time series SARIMA model can be applied as an efficient tool to forecast the energy needs to ensure a long-term development and energy management in the building sector.