Modeling solar extinction using artificial neural networks. Application to solar tower plants. (15th May 2020)

• Record Type: Journal Article
• Authors: Ballestrín, J.; Carra, E.; Alonso-Montesinos, J.; López, G.; Polo, J.; Marzo, A.; Fernández-Reche, J.; Barbero, J.; Batlles, F.J.
• Is Part Of: Energy. Volume 199(2020)
• Publication Date: 2020-05-15
• Subjects: Solar extinction -- Concentrated solar power (CSP) -- Solar tower plants -- Artificial neural networks (ANN); Power resources -- Periodicals; Power (Mechanics) -- Periodicals; Energy consumption -- Periodicals; 333.7905
• DOI: 10.1016/j.energy.2020.117432
• Languages: English
• ISSNs: 0360-5442

Abstract

Abstract: The extinction of solar radiation is considered an important variable to take into account in the design and operation of commercial concentrated solar power (CSP) tower plants, where the distances between the concentrating heliostats and the receiver on top of the tower are, in many cases, around 1 km. Aerosol particles and water vapor in the path traveled by solar radiation are the main causes of its extinction due to the phenomena of scattering and absorption. Since June 2017, solar extinction has been reliably measured daily at Plataforma Solar de Almería, which has allowed analyzing the dependence of this parameter with other meteorological variables. It has been observed that during high turbidity events there is a clear linear dependence of the solar extinction with the particle concentration and humidity. This work shows that, although this linear character is diluted under normal conditions, artificial neural networks (ANN) allow modeling and predicting extinction as a function of these two magnitudes. Highlights: Artificial neural networks (ANN) allow modeling solar extinction. Dependence of the solar extinction with the particle concentration and humidity. ANN models for solar tower plants.