Weather Uncertainty Data for Stochastic Optimization [back]

   In this website you will find a collection of data sets containing weather uncertainty information generated from the numerical weather predicition model WRF and Gaussian Process models.  These data sets can be used in diverse stochastic optimization studies.  Since a lot of effort has been dedicated to collecting this information, we kindly ask you to cite the reference papers if you use the data.

Pittsburgh Ambient Temperature Samples
  • Setting: August 1st-6th, 2006 running WRF with a 5 day forecast horizon.  
  • Applications:  Building Energy Management, Power Plant Optimization, Load Forecasting.
  • Reference:  Zavala, V. M.; Constantinescu, E. M.; Krause, T.; and Anitescu, M.  On-Line Economic Optimization of Energy Systems Using Weather Forecast Information. Journal of Process Control, 19(10), pp. 1725-1736, 2009. [pdf]

Illinois Wind Speed and Power Samples [link]
  • Setting: June 2nd-4th, 2006 running WRF with a 24 hr forecast horizon and reassimilating data every 24 hr. 
  • Applications:  Wind Forecasting for Unit Commitment and Economic Dispatch.
  • Reference:  Constantinescu, E. M.; Zavala, V. M.; Rocklin, M.; Lee, S.; and Anitescu, M.  A Computational Framework for Uncertainty Quantification and Stochastic Optimization in Unit Commitment with Wind Power Generation.  IEEE Transactions on Power Systems, 26(1), pp. 431-441, 2010. [pdf]