Development of Real-Time Hotspot Detection System Utilizing Artificial Intelligence in PV Generation System

Abstract

Photovoltaic (PV) Generation system is one of the easiest renewable energy systems to generate either small amounts of energy for usage in households or for large amounts as utilized in fields. Although PV generation system does not burn fuel for power generation, some problems exist regarding heat. One of these problems is called Hotspots. A Hotspot is an increase in the cell`s heat in certain conditions and positions. In some cases, the heat can even ignite a fire. In this study, we propose a new method to detect this hotspot phenomenon at an early stage. The proposed method utilizes Artificial Intelligence (AI) as the main detection system. In fact, we were able to detect the hotspot with an accuracy of 82.25% using only two parameters, string current and string voltage. This system is a secondary system to be used in conjunction with the main control system. The output will be a flag sent to the main controlling system. Designing this system as secondary one, makes it easier to apply in already constructed PV fields. The findings illustrated the detection of hotspots with an accuracy rate of 82.25% using only two parameters, namely string current and string voltage. Thus the findings from this study provides a basis for the future development of a system which provides an overall evaluation for solar panels including hotspots and degradation.