Feature extraction from infrared sky images for solar energy estimation

In recent years, the shift towards renewable energy sources, particularly solar energy, has gained momentum due to its environmental benefits. Maximizing solar energy production requires accurate information on atmospheric conditions, notably incident solar radiation. While satellite images have traditionally been used for this purpose, the emergence of sky cameras offers a promising technology for studying sky conditions with higher precision. This paper aims to extract features from panoramic sky images captured with an infrared camera to estimate intra-hour solar radiation. Leveraging classical techniques from image analysis and computer vision such as segmentation and optical flow algorithms, in conjunction with specialized photovoltaic energy systems libraries such as pvlib, a unified pipeline is developed to extract relevant data for input into solar prediction models. The integration of satellite images and sky camera data, combined with these sophisticated techniques, has shown effectiveness in short-term cloud prediction. This facilitates better adaptation of solar power plant operations to meteorological conditions and grid integration.

• Autores: Hernández Vanegas, Rodrigo
• Tipo de recurso: Trabajo de grado de pregrado
• Fecha de publicación: 2024
• Institución: Universidad de los Andes
• Repositorio: Séneca: repositorio Uniandes
• Idioma: eng
• OAI Identifier: oai:repositorio.uniandes.edu.co:1992/74511
• Acceso en línea: https://hdl.handle.net/1992/74511
• Palabra clave: Optical Flow; Image Segmentation; Solar Power Forecasting; Feature Extraction; Infrared Sky Camera; Ingeniería
• Rights: openAccess
• License: Attribution 4.0 International