Support vector machine based maximum power degradation detection in photovoltaic modules
|
Full Text |
Pdf
|
|
Author |
O. Hemakesavulu, Pradyumna Kumar Dhal, Shaik Hussain Vali, Sadhu Radha Krishna, P. Yamuna and Vempalle Rafi
|
|
e-ISSN |
1819-6608 |
|
On Pages
|
117-124
|
|
Volume No. |
21
|
|
Issue No. |
2
|
|
Issue Date |
March 20, 2026
|
|
DOI |
https://doi.org/10.59018/012622
|
|
Keywords |
deterioration of photovoltaic, support vector machine.
|
Abstract
During five months of operation in a moderately humid atmosphere, this article uses a Support Vector Machine (SVM) classifier to examine the deterioration of photovoltaic (PV) modules in India. There is a maximum nominal power of 270 W for the PV module that was checked. An extensive electrical and environmental dataset was used by the SVM model to successfully categorize the module's performance statuses. During peak irradiation hours, the module operated under cloudy conditions for nearly 30% of the time, leading to noticeable power fluctuations and contributing to degradation effects. According to the analysis of irradiance evolution over time, the peak power delivered during the day reaches about 249 W when solar irradiance ranges from 950 W/m² to 1050 W/m², which is about 92% of the module's nominal power. Heat maps that intuitively show deterioration trends were created using SVM-based classification, providing a diagnostic tool that is clearer and more interpretable than standard analytical approaches. Supporting better monitoring, maintenance, and performance optimization in comparable climatic situations, the findings show that the suggested technique is efficient for identifying deterioration in individual PV modules and may be scaled to PV power plants.
Back