DOI: 10.46810/tdfd.1914730 ISSN: 2149-6366

Simulation and Estimating of Monthly Energy Production of Solar Power Plant Using PVsyst Program and Advanced Deep Learning Algorithms

Mehmet Çınar, İrfan Ökten
Show PDF Cite
Solar energy production forecasting is crucial for optimally integrating renewable energy sources into power systems. Deep learning techniques have emerged as promising alternatives for solar energy forecasting in recent years.In this study; Modeling, simulation and estimation of solar energy that can be produced next year of the solar power plant with a total installed power of 1300 kW, established in Bitlis province in the Eastern Anatolia Region of Turkey, are shown. PVsyst software program was used to analyze the performance ratio and different losses occurring in the system. Additionally, the study reviews deep learning techniques for solar energy forecasting with a special focus on Long Short Term Memory (LSTM) networks for short-term solar energy production forecasting. As a result of this examination, R2 score, MAE (Mean Absolute Error), and MSE (Mean Squared Error) were obtained as 0.863, 0.284 and 0.125, respectively. Estimated energy production in 2024 was calculated from the network trained with the LSTM model. Actual results were interpreted by comparing the results of the PVsyst program and the artificial intelligence algorithm. The study provides high-accuracy prediction to improve the integration of solar energy into power systems and reduce energy costs

More from our Archive