Best Practices Handbook for the Collection and Use of Solar Resource Data for Solar Energy Applications

Abstract

ForewordThe first version of this handbook was developed in response to a growing need by the solar energy industry for a single document addressing the key aspects of solar resource characterization. The solar energy industry has developed rapidly throughout the last few years, and there have been significant enhancements in the body of knowledge in the areas of solar resource assessment and forecasting. Thus, this second version of the handbook was developed from the need to update and enhance the initial version and present the state of the art in a condensed form for all of its users.Although the first version of this handbook was developed by only researchers from the National Renewable Energy Laboratory, this version has additional contributions from an international group of experts primarily from the knowledge that has been gained through participation in the International Energy Agency's Solar Heating and Cooling Programme Task 36 and Task 46.As in the first version, this material was assembled by scientists and engineers who have many decades of combined experience in atmospheric science, radiometry, meteorological data processing, and renewable energy technology development.Readers are encouraged to provide feedback to the authors for future revisions and an expansion of the handbook's scope and content. iv This report is available at no cost from the National Renewable Energy Laboratory (NREL) at www.nrel.gov/publications. PrefaceAs the world looks for low-carbon sources of energy, solar power stands out as the single most abundant energy resource on Earth. Harnessing this energy is the challenge for this century. Photovoltaics, solar heating and cooling, and concentrating solar power (CSP) are primary forms of energy applications using sunlight. These solar energy systems use different technologies, collect different fractions of the solar resource, and have different siting requirements and production capabilities. Reliable information about the solar resource is required for every solar energy application. This holds true for small installations on a rooftop as well as for large solar power plants. However, solar resource information is of particular interest for large installations, because they require a substantial investment, sometimes exceeding $1 billion in construction costs. Before such a project is undertaken, the best possible information about the quality and reliability of the fuel source must be made available. That is, project developers need to have reliable data about the solar resource available at specific locations, including historic trends with seasonal, daily, hourly, and (preferably) subhourly variability to predict the daily and annual performance of a proposed power plant. Without this data, an accurate financial analysis is not possible.In September 2008, the U.S. Department of Energy (DOE) hosted a meeting of prominent CSP developers and stakeholders. One purpose was to identify areas in which the DOE's CSP program should focus its efforts to help the industry develop an...

Solar Forecasting in a Challenging Insular Context

“…Post-processing methods like Model Output Statistics (MOS) are frequently applied to reduce these biases or to refine the output of NWP models, as detailed local weather features are generally not resolved by NWP predictions [3]. In addition, spatial averaging can contribute to reduce forecast errors, particularly in situations with variable clouds, that are difficult to predict ( [4,17]). To our best knowledge, these post-processing techniques have been only tested in a continental context ( [3][4][5]) and it appears interesting to assess in this survey the performances of these techniques in the case of an insular environment.…”

Comparison of satellite-retrieved high-resolution solar radiation datasets for South Africa

“…The mean bias difference (MBD), the root mean square difference (RMSD), and mean absolute difference (MAD) in absolute (Wm -2 ) and relative (%) values (rMBD, rRMSD and rMAD) were calculated according to Equations 1 to 6 [3,7,24]. In addition to these, the R 2 correlation coefficient was also calculated using Equation 7 [7,24]. …”

“…When no on-site measurements are available, other techniques should be applied, such as empirical models based on meteorological variables like temperature or sunshine duration [3,4] or models based on reanalysis and retrospective weather prediction models [5,6]. Satellite-based models have become a very powerful tool for estimating the solar resource at high and uniform spatial resolution (typically a few kilometres) and temporal resolution (hourly or better) over large geographical areas [3,7]. Satellite-based estimates have been validated in the scientific literature [8][9][10][11][12].…”

A clustering approach for the analysis of solar energy yields: A case study for concentrating solar thermal power plants

“…Sunlight is the fuel for all solar energy generation technologies, and knowledge of the quality and future reliability of this fuel is essential to accurate analyses of system performance and the financial viability of a project 1,2 . In particular, a precise knowledge of the incoming Direct Normal solar Irradiance (DNI) is required for an accurate design and operation of Concentrating Solar Thermal Power (CSTP) plants 3,4 .…”