Predicting the renewable energy consumption in 2026 by using a recursive moving average model

Abstract

Due to the growing use of renewable energy resources as a sustainable solution for supplying energy, forecasting of consumption and production in energy sector and identifying the effective policies, infrastructures and technologies is vital for reaching the global aims. Reaching the more accurate solutions needs to consider more variables or more complex models. This is not easy to use in many cases due to lacking of the accessible data or low understanding of the relations between parameters. In this study, a recursive moving average model as a black-box approach was used to predict renewable energy consumption on the global scale. The model was trained by using data from 1990 to 2017 until the output of the model reached the desired accuracy. Then the calibrated model was used to predict renewable energy consumption until 2026. Finally, the mean absolute error (MAE) of the results was evaluated for the obtained results. The results showed that the proposed model was able to predict the amount of renewable energy consumption with 95.79% accuracy in a period of 9 years with the MAE of 0.043. Also, the results showed that the target parameter seems to be increased by 9.01% from 2018 to 2026 annually.

KEYWORDS:

• Predictive model
• recursive relation
• moving average
• renewable energy consumption

Disclosure statement

No potential conflict of interest was reported by the author(s).