How Solar Thermal Collector Performance Was Modeled



  • Climate Data:
    • Data is made available by the National Solar Radiation Database (NSRDB) for 858 weather stations in the United States of America.
    • Meteorological data in the NSRDB is provided by the National Climate Data Center (NCDC).
    • Solar radiation data in the NSRDB is generated by models developed by the National Renewable Energy Laboratory (NREL).
      • Primary model used was the Meteorological-Statistical (METSTAT) model.
      • All solar radiation data is modeled, meaning it was not measured, but rather, calculated based on historical measured meteorological data.
  • Modeling Inputs:
    • Geographical location of collectors; longitude and latitude
    • Collector altitude and azimuth angle (fixed collector tilt)
    • Collector surface area (m2)
    • Collector performance characteristics obtained from third party test results (such as Universität Stuttgart Institut für Thermodynamik (ITW)).
    • Model assumes no shading or obstructions of the collector.
  • Performance characteristics of possible components used in the system configuration
    • Solar controller: Pump speed varied such that collector minimum output temperature is controlled to match component input temperature.
    • Pump(s): Minimum power consumption = 9 Watts (if collector power output is less than 9W the pump is assumed to not turn on, resulting in no power collection).
    • Piping: Heat loss = 0 Watts (can't predict how you will configure the rest of your system so you must subtract this power loss yourself).
  • Solar radiation, sun position, and dry-bulb temperature data is used from the NSRDB for each hour between 1991-2005 at the specified geographic location (leap days skipped).
  • Solar radiation (direct + diffuse) available on a tilted collector surface is calculated using NSRDB data and methods developed by Duffie and Beckman.
    • Direct
      • Data for direct beam radiation on a horizontal surface is converted to direct beam radiation on a tilted collector surface.
      • Incident angle modifiers, both translational and longitudinal values are interpolated from test results and multiplied times the direct beam radiation.
    • Diffuse
      • Data for sky diffuse radiation on a horizontal surface is converted to diffuse radiation on a tilted surface using the view factor from principles of heat and mass transfer.
      • Surface reflected diffuse radiation incident upon a tilted surfcace is calculated using view factors, incident solar radiation on a flat surface, and the surface albedo (percentage incident radiation which is diffusely reflected) provided by the NSRDB (which varies with season and location).
      • Diffuse incident angle modifiers are currently assumed to be 1.0 (this results in a slightly over approximated solar collector power output.
  • Using collector performance values, weather conditions, plans and system characteristics, and available solar radiation, power output of the collector (kW) is calculated for each hour.
  • From this simulation several other values are derived to better understand the collector array performance:
    • Energy collected per day (kWh/day) for each day between 1991 and 2005
    • Max/Median/Min energy collected per day (kWh/day)
    • Efficiency of the collectors to convert available solar radiation into usable heat energy
  • Because each hour is modeled the data can be used to represent extreme conditions as well as average conditions. This can be used for properly sizing energy storage and collector arrays for critical systems where solar may be the only power source.
  • Additional Notes:
    • Because solar radiation is a variable power source, the process of sizing a solar thermal system also involves choosing a solar fraction, and defining the probability of achieving that solar fraction.
      • Solar fraction represents the minimum percentage of power demands satisfied by solar power.
      • Due to the uncertainty of available solar power, designing for a 100% solar fraction may result in a system being over sized for a large percentage of the time.
    • Historical solar collector power output has been provided for reference purposes only. We have done our best to simulate solar collector power output, but as in any mathematical model, there will be some assumptions, data collection errors, unquantified variation, etc. which may cause actual collector power output to vary from the simulated output.
    • The simulation indicates how a solar collector would have performed in the past. Due to the variable nature of the weather in any location in the world, it is impossible to predict future weather conditions. Using this historical simulation, it is the system designers responsibility to make an engineering judgement how the solar collector will perform in the future.



©2007 - Andy Schroder