2020/12/02 by Quentin Paletta, Paletta, Quentin, Joan Lasenby +1
Computer Science · Energy · #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Image Enhancement Techniques #Machine Learning (cs.LG) #Photovoltaic System Optimization Techniques #Solar Radiation and Photovoltaics
paper · pdf · doi:10.48550/arxiv.2012.01059
openalex publication_date 2020/12/02 · openalex created_date 2022/11/01 · openalex updated_date 2026/07/28
Improving irradiance forecasting is critical to further increase the share of\nsolar in the energy mix. On a short time scale, fish-eye cameras on the ground\nare used to capture cloud displacements causing the local variability of the\nelectricity production. As most of the solar radiation comes directly from the\nSun, current forecasting approaches use its position in the image as a\nreference to interpret the cloud cover dynamics. However, existing Sun tracking\nmethods rely on external data and a calibration of the camera, which requires\naccess to the device. To address these limitations, this study introduces an\nimage-based Sun tracking algorithm to localise the Sun in the image when it is\nvisible and interpolate its daily trajectory from past observations. We\nvalidate the method on a set of sky images collected over a year at SIRTA's\nlab. Experimental results show that the proposed method provides robust smooth\nSun trajectories with a mean absolute error below 1% of the image size.\n