ELECTRICAL CHARACTERISTIC CLASSIFICATION OF THE PV’S USING SUPPORT VECTOR MACHINES

Authors : Fatih SERTTAŞ, Fatih Onur HOCAOĞLU

Pages : 199-208

Doi:10.18038/estubtda.901800

View : 18 | Download : 10

Publication Date : 2021-06-29

Article Type : Research Paper

Abstract :Photovoltaics have more attraction day by day due to increasing technologies and decreasing prices. However, in practice, the applicants encounter fundamental problems such as shadow effects and degradation. Due to these problems, the amount of produced annual electricity decreases considerably. Moreover, it is not easy to understand if the company`s PV satisfies the requirements or is appropriate for the application. In order to solve such problems, it is instructed to examine the electrical characteristics of the modules. An initial task should be the classification of PV modules according to performance results. Here in this study, it is aimed to classify different PV modules, including different output characteristics. It is aimed to show whether it is possible to classify different PV modules including the same output power or not? To find the answer to this question, a test platform is built. 4 different panels are tested on the platform. While the test, the panels produced by different companies split into two groups, each has the same output power. Under a different insulation condition, the test is performed, and the cells` current-voltage curves are constructed. Different statistics are extracted by using this information. Different variations of these statistics are presented to multi-SVM. Finally, accurate classification results are obtained.
Keywords : Photovoltaic, IV Curve, Classification, Electrical Characteristics, Support Vector Machines