Por favor, use este identificador para citar o enlazar este ítem: http://hdl.handle.net/10259/4752
Título
Diffuse solar irradiance estimation on building's façades: review, classification and benchmarking of 30 models under all sky conditions
Publicado en
Renewable and Sustainable Energy Reviews. 2017, V. 77, p. 783-802
Editorial
Elsevier
Fecha de publicación
2017-09
ISSN
1364-0321
DOI
10.1016/j.rser.2017.04.034
Resumen
Solar energy potential analysis on oriented and tilted surfaces, such as building façades, is of capital importance to estimate the energy production potential in integrated systems, such as BIPVs. The spatial distribution of solar diffuse energy outstands as one of the most critical factors in order to improve performance simulations. Nevertheless, most wide spread models have been only evaluated on equator facing tilted surfaces and with daily mean or hourly mean time ranges. Thus, in this paper, 30 transposition solar diffuse irradiance models, from the semiphysical classical ones to the newest non-parametric models, are reviewed, classified according to their characteristics and evaluated against empirical 10-min averaged diffuse irradiance values gathered from high precision pyranometers placed on vertical positions facing the four cardinal directions. Models’ performance is evaluated by several statistical estimators and a benchmark has been carried out by a non-parametrical aggregating procedure. Results show that the most accurate models appear to be the non-parametric ones. From these, The Multi-Layer Perceptron obtains the best results. From the parametric models, the one whose estimations are closest to the measures is the Perez et al. with local optimized coefficients. Perez et al. model with the original coefficients, Skartveit and Olseth and Igawa et al. models also show a good performance.
Palabras clave
Diffuse solar irradiance
Vertical surfaces
Irradiance modelling
Sky conditions
Benchmarking
Materia
Electrotecnia
Electrical engineering
Versión del editor
Aparece en las colecciones
Documento(s) sujeto(s) a una licencia Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International