Long Term Forecasting of Electrical Energy Consumption for Iran Based on Optimized Artificial Neural Networks and Socio-Economic Indicators Data

The objective of this study is to forecast the electrical energy consumption (EEC) of Iran from 2023 to 2050 based on the historical data from 1967 to 2022. To meet the objective, first, the total energy consumptions in the residential, commercial, industrial, transportation and agricultural sectors are forecasted using machine learning algorithms. To forecast the EEC, socio-economic indicators including population and gross domestic product (GDP), level of export and import of energy and total energy consumption in five sectors have been used as input of an artificial neural network (ANN). The architecture of ANN is optimized using particle swarm optimization (PSO) algorithm. The forecast results show a 2.84 time increase in EEC for Iran in 2050 to achieve the needed growth for human development index (HDI).