Statistical Modeling for Improvement of Numerical-Model-Based Solar Radiation Forecasts
We first analyze some features of numerical weather predictions (NWP) for global solar radiation and notice that they are undersmooth. This finding opens a way to improvements via various smoothing strategies. Then we introduce a statistical modeling framework based on modern semiparametric regression. We use a numerical weather prediction (NWP) model output as one of the inputs for our statistical model. The statistical model is build on the modern regression formalism, utilizing nonparametric B-splines for nonlinear parts whose exact shape is unknown a priori (apart from physically motivated smoothness). Then we illustrate its abilities for systematic development of strategies for NWP calibration and further development. The results are useful both for practical forecasting and as a source of feedback for NWP modelers.
KeywordsGeneralized additive model Numerical weather forecast model Global solar radiation Calibration Semiparametric modeling
The work on this article was partly supported by the long-term strategic development financing of the Institute of Computer Science (RVO:67985807) and by the Czech Science Foundation grant GA13-34856S, Advanced random field methods in data assimilation for short-term weather prediction.
- 1.EPIA 2014: European Photovoltaic Industry Association. Global Market Outlook for Photovoltaics 2014–2018. http://www.cleanenergybusinesscouncil.com/site/resources/files/reports/EPIA_Global_Market_Outlook_for_Photovoltaics_2014-2018_-_Medium_Res.pdf
- 2.COST Action ES1002: Weather Intelligence for Renewable Energies. http://www.wire1002.ch/
- 3.Diagne, H.M., David, M., Lauret, P., Boland, J.: Solar irradiance forecasting: state-of-the-art and proposition for future developments for small-scale insular grids. In: Proceedings of the World Renewable Energy Forum 2012, Denver, USA (2012)Google Scholar
- 6.Kumar, R., Aggarwal, R.D., Sharma, J.D.: New regression model to estimated global solar radiation using artificial neural network. Adv. Energy Eng. 1, 66–73 (2013)Google Scholar
- 8.Warner, T.T.: Weather Modeling and Forecasting of PV System Operation. Cambridge University Press, Cambridge (2011)Google Scholar
- 15.R Core Team 2016. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. http://www.R-project.org/
- 16.The R-INLA project 2016. http://www.r-inla.org/