Estimating of solar radiation with artificial neural networks and genetic algorithms and comparison of performances: Adana province example

Fossil fuels, which have been considered as a solution to the problem of increasing energy demand in parallel with the population for many years, have begun to leave their place to renewable energy sources due to their unaffordable costs and the irreversible damage they cause to the environment. Considering the sustainability, environmental friendliness, and potential of satisfying the energy deficit of today's world, these resources continue to gain popularity. Among the renewable energy sources, solar energy-based investments, which is one of the most preferred sources due to the geographical advantage of our country, and its utilization in our daily lives are ascending. Global solar radiation is one of the most important parameters related to solar energy, and its measurement is quite laborious and costly. Considering these disadvantages, estimation and modeling options gain importance under the roof of studies related to solar energy. In this study, global solar radiation estimation has been achieved through Artificial Neural Networks (ANN) and Genetic Algorithms (GA) algorithms in the MATLAB environment, using the hybrid data set for the years 2015-2021 for Adana province. In addition, a performance comparison is given by considering the coefficient of determination of these two algorithms. As a result of the study, it has been observed that the GA algorithm performed a more successful result compared to the ANN algorithm in estimating the global solar radiation intensity.