Source code 

Model Identifier and version number

  • solar_power 
  • Version 1.0

Citation Information

  • Title
    • DOE Solar Power Conversion
  • Abstract
    • Reads forecast irradiance input for all solar farm sites. Applies static data mining models developed for each farm to convert forecast irradiance to forecast power.
  • Author(s)
    • Gerry Wiener, Julia Pearson, Tom Brummet
  • Point of Contact
    • Tom Brummet
  • Creation Date
    • February 29, 2016
  • Modification Date
  • Identifier Code
    • DOE Solar Technology Transfer
  • Use Constraints
    • None

Distribution Information

 

 

Model description

Intended use

  • To produce power forecasts using forecast irradiance values for cost share utility partners

Key assumptions

  • Requires solar irradiance forecast at the location of interest in order to produce power forecast

Documentation and References

  • Installation documentation including hardware and software requirements
  • Data mining approach references
    • Information on the Cubist regression tree algorithm:
      • Quinlan. Learning with continuous classes. Proceedings of the 5th Australian Joint Conference On Artificial Intelligence (1992) pp. 343-348 
      • Quinlan. Combining instance-based and model-based learning. Proceedings of the Tenth International Conference on Machine Learning (1993a) pp. 236-243
      • Quinlan. C4.5: Programs For Machine Learning (1993b) Morgan Kaufmann Publishers Inc. San Francisco, CA 
      • Wang and Witten. Inducing model trees for continuous classes. Proceedings of the Ninth European Conference on Machine Learning (1997) pp. 128-137
    • Information on using Cubist in converting wind to power:
      • William P. Mahoney, Keith Parks, Gerry Wiener, Yubao Liu, William L. Myers, Juanzhen Sun, Luca Delle Monache, Thomas Hopson, David Johnson, and Sue Ellen Haupt, "A Wind Power Forecasting System to Optimize Grid Integration," in IEEE Transactions on Sustainable Energy, Vol. 3, No. 4, October 2012, pg 670-681.
  • No labels