Using Matlab real-time image analysis for solar panel fault detection with UAV

Abstract

The main purpose of this study is to evaluate the feasibility to use Unmanned Aerial Vehicle (UAV) technology for solar panel applications and to propose a reliable, economical and fast method of fault detection. In this experiment, the UAV was used to inspect the panels of a solar farm. For this purpose, a set of tools were selected including an infrared camera and a High Definition (HD) photographic lens to scan solar panel assembly. The infrared images were immediately transmitted back with backhaul HD for analysis. The infrared image display can be roughly divided into three health conditions. The first one is the normal operation of the solar panel. In the second case, the infrared image represents an abnormal situation of the solar panel. The third one is a blurred case in which the infrared image cannot be determined to be normal or abnormal. Under this kind of situation, the Matlab® image analysis was used in this study to determine the health condition of solar panels. The results show that it is effective and efficient to use UAV to capture the thermal image and to detect different faults of solar modules. Moreover, this process is fast, low cost, and can detect defects that are not noticeable by visible light. It may provide maintenance personnel with a tool to prevent solar module failure.