Abstract
Today, with the intensive use of electrical devices, the need for electricity has increased. Fossil fuels are generally used to meet this need. However, to reduce the damage caused by fossil fuels to the environment, governments offer various incentives to renewable energy sources. In this context, incentives for solar power plants are also quite high. Recently, many investors wanted to establish a solar power plant. Since establishing a solar power plant requires high investment costs, the calculation of amortization periods plays an important role in making investment decisions. The development of technology has made it possible to predict the amortization times of these costs with artificial intelligence algorithms. In this study, energy data that can be produced in the future have been estimated with deep learning algorithms using real solar power plant data. The data were taken from solar power plants belonging to Humartaş Energy company. In the study, analyzes and predictions were made using the LSTM (Long Short-Term Memory) method, which is used extensively in time series algorithms. The error rate of the study was found to be between 1% and 17%. It is predicted that this study can be used in other renewable energy sources such as wind, hydraulic, geothermal energy.