Application of the genetic algorithm to the estimation of energy consumption in Poland

Abstract

This work aims to model and estimate electricity consumption in Poland using a genetic algorithm approach. The genetic model was based on socio-economic data such as GDP, imports and exports, and population growth. The data covered the period from 1990 to 2024. Quadratic and exponential models were tested in this study. The exponential model was demonstrated to be the best fit to the data. The best mean absolute error was obtained for the exponential model, MAE = 0.0276, for a population size of 100, a crossover probability of 0.85, and a mutation probability of 0.02. The projection of electricity consumption in Poland was also demonstrated and compared with the data from the literature. The genetic model forecasts an electricity consumption of 211 TWh in 2040.