Dronalyse: Cloud-based Drone for Photovoltaic Fault Inspection and Analysis

Photovoltaic cells otherwise known as solar cells convert sunlight into electricity. However, there are militating issues against optimal performance of this technology. Traditional methods of resolving these issues have not yielded satisfactory results. This study aims to resolve the abovementioned issues using a novel Dronalyse, a drone-based method for detecting and analyzing photovoltaic faults. The Tshwane University of Technology, Soshanguve South Campus, South Africa hosted the data collection team; DJI Mavic 3 Thermal's high-resolution thermal and visual imaging capabilities were used for the data collection. The data collected comprised Hotspots images, Cracks images, and Shadings images. The following services were employed to perform the tasks; AWS IoT Core, Amazon S3, AWS Lambda, Amazon Rekognition, OpenCV on Lambda, Amazon SageMaker, Amazon QuickSight, AWS SNS (Simple Notification Service), and AWS DynamoDB Database. Extensive analysis of the thermal data reveals the anomalies as indicative of faults in the solar cells of PV module. The data we collected were processed and shared freely for scientific community use at https://doi.org/10.17632/5ssmfpgrpc.l