Analysis of the Performance of Different Types of PV Panels in Spring and Summer Using Regression Methods
2025
Salim Kılıç | Ertuğrul Adıgüzel | Erkan Atmaca
The present study employs machine learning regression analyses to investigate the efficiency of photovoltaic (PV) panels utilizing solar energy under the influence of environmental factors. The experimental study was conducted on two 100-watt monocrystalline and two polycrystalline PV panels, which were divided into clean and dirty groups. The following variables were monitored and recorded for a period of six months: radiation, panel temperature, air temperature, wind speed, humidity, pressure, and ultraviolet (UV) radiation. Additionally, current, voltage, and power were recorded. These measurements were taken during the production of energy by PV panels. Monocrystalline PV panels exhibited an 8.6% increase in energy efficiency, while polycrystalline PV panels demonstrated a 6.2% increase, following the collection and cleaning of data in accordance with the IEC 61724 standard. Six distinct machine learning regression analyses were conducted on the dataset. The results were compared using the Root Mean Square Error (RMSE) and the coefficient of determination (R<sup>2</sup>). The Random Forest model demonstrated the greatest predictive success, while the Support Vector Regression (SVR) model exhibited the lowest performance.
Show more [+] Less [-]AGROVOC Keywords
Bibliographic information
This bibliographic record has been provided by Directory of Open Access Journals