Realization of Turkey’s energy demand forecast with the improved arithmetic optimization algorithm

Abstract

Due to the increasing energy consumption, energy has become a constant problem in the world. Rapidly increasing population, urbanization and economic activities increase the pressure of countries on energy. In a world where consumption is increasing, energy management has become a more important and challenging issue. For this reason, it is necessary to make proper estimations that will reduce the pressure of energy demand on this issue. In order to realize the estimation of energy demand, Turkey application is carried out in this study and arithmetic optimization algorithm (AOA) which is a stochastic metaheuristic algorithm has used to for solving energy demand problem. AOA is inspired from four substantial math functions such as subtraction, multiplication, addition and division for searching process of candidate solutions. The current position update rules of AOA are not powerful enough for solving the problem dealt with this study. Therefore, an improved version of AOA named as IAOA is proposed for solving energy demand problem. In the proposed algorithm, a new position update rule is incorporated to basic AOA in order to enhanced the exploration and exploitation capability of AOA. The linear regression model is used for the estimation of the energy demand and the population, domestic product, import and export data are used in estimation process. In the proposed model, Turkey’s real data samples for the years 1979–2011 have been used, and Turkey’s long-term energy demand has been estimated for the years 2012–2030. While performing the estimation process, Turkey’s energy data of the years 1979–2011 have processed, and then Turkey’s long-term energy demand estimations are realized for three different scenarios. Firstly, the experimental results of the proposed model are analyzed, then the results are compared with different studies proposed in the literature. As a result of the comparisons, it is seen that the IAOA method has achieved better or similar results than compared methods. For this reason, it can be said that the IAOA method is competitive and successful in realizing the energy demand forecast for Turkey’s future years.