摘要：The dynamic reconfiguration and maximum power point tracking in large-scale pho-tovoltaic (PV) systems require a large number of voltage and current sensors. In particular, thereconfiguration process requires a pair of voltage/current sensors for each panel, which introducescosts, increases size and reduces the reliability of the installation. A suitable solution for reducingthe number of sensors is to adopt image-based solutions to estimate the electrical characteristics ofthe PV panels, but the lack of reliable data with large diversity of irradiance and shading conditionsis a major problem in this topic. Therefore, this paper presents a dataset correlating RGB imagesand electrical data of PV panels with different irradiance and shading conditions; moreover, thedataset also provides complementary weather data and additional image characteristics to supportthe training of estimation models. In particular, the dataset was designed to support the design ofimage-based estimators of electrical data, which could be used to replace large arrays of sensors. Thedataset was captured during 70 days distributed between 2020 and 2021, generating 5211 images andregisters. The paper also describes the measurement platform used to collect the data, which willhelp to replicate the experiments in different geographical locations.
关键词：photovoltaic;image-based estimation;partial shading;current vs. voltage characteristic
