Interpolating and Estimating Horizontal Diffuse Solar Irradiation to Provide UK-Wide Coverage: Selection of the Best Performing ModelsReportar como inadecuado


Interpolating and Estimating Horizontal Diffuse Solar Irradiation to Provide UK-Wide Coverage: Selection of the Best Performing Models


Interpolating and Estimating Horizontal Diffuse Solar Irradiation to Provide UK-Wide Coverage: Selection of the Best Performing Models - Descarga este documento en PDF. Documentación en PDF para descargar gratis. Disponible también para leer online.

Centre for Renewable Energy Systems Technology, School of Mechanical, Electrical and Manufacturing Engineering, Loughborough University, Loughborough, Leicestershire LE11 3TU, UK





*

Author to whom correspondence should be addressed.



Academic Editor: Senthilarasu Sundaram

Abstract Plane-of-array PoA irradiation data is a requirement to simulate the energetic performance of photovoltaic devices PVs. Normally, solar data is only available as global horizontal irradiation, for a limited number of locations, and typically in hourly time resolution. One approach to handling this restricted data is to enhance it initially by interpolation to the location of interest; next, it must be translated to PoA data by separately considering the diffuse and the beam components. There are many methods of interpolation. This research selects ordinary kriging as the best performing technique by studying mathematical properties, experimentation and leave-one-out-cross validation. Likewise, a number of different translation models has been developed, most of them parameterised for specific measurement setups and locations. The work presented identifies the optimum approach for the UK on a national scale. The global horizontal irradiation will be split into its constituent parts. Divers separation models were tried. The results of each separation algorithm were checked against measured data distributed across the UK. It became apparent that while there is little difference between procedures 14 Wh-m2 mean bias error MBE, 12 Wh-m2 root mean square error RMSE, the Ridley, Boland, Lauret equation a universal split algorithm consistently performed well. The combined interpolation-separation RMSE is 86 Wh-m2. View Full-Text

Keywords: photovoltaic; spatial interpolation; ordinary kriging; solar radiation separation; national model validation; UK photovoltaic; spatial interpolation; ordinary kriging; solar radiation separation; national model validation; UK





Autor: Diane Palmer * , Ian Cole, Tom Betts and Ralph Gottschalg

Fuente: http://mdpi.com/



DESCARGAR PDF




Documentos relacionados