â The present invention automatically detects and merges tovoltaic panels taken based on CAD drawings to automatically detect defective panels, thereby providing an innovative technique to increase the accuracy of the diagnosis system and reduce inspection time by applying AI.â In the present invention, it is possible to take automatic flight togra by setting the flight path of the drone using the CAD drawing for installing the solar array.â The drone takes a high-definition thermal image of the panel of the solar power plant in block units while flying at a certain altitude using the set automatic flight path.â The captured high-definition thermal image is overlapped with the CAD design drawing of the solar power plant to generate image information for solar array diagnosis.â The generated diagnostic image uses an AI diagnostic algorithm to diagnose the abnormality of the solar array.â The present invention allows a large-scale solar power plant to be managed quickly and efficiently with a small of managers.