View Source

Title
- DOE Solar StatCast (RD-ANN)
Abstract
- This regime dependent artificial neural network (RD-ANN) system classifies cloud regimes with a k-means algorithm based on a combination of surface weather observations, irradiance observations and GOES-East satellite data. The ANNs are then trained on each cloud regime to predict the clearness index.
Author(s)
- Tyler McCandless, Tom Brummet, Sue Dettling, Sue Haupt
Point of Contact
- Tom Brummet
Creation Date
- September 27, 2016
Modification Date
Identifier Code
- DOE Solar Technology Transfer
Use Constraints
- None

This regime dependent artificial neural network (RD-ANN) system classifies cloud regimes with a k-means algorithm based on a combination of surface weather observations, irradiance observations and GOES-East satellite data. The ANNs are then trained on each cloud regime to predict the clearness index.

To produce irradiance forecasts using previous irradiance values, along with weather variables.

Reqiures solar irradiance observations at the location of interest and nearby weather observations taken at the surface and satellite based observations in order to produce irradiance forecast.Reqiures solar irradiance observationss at the location of interest and nearby weather forecasts in order to produce irradiance forecast.

Processing documentation including hardware and software requirements
- RDANN_DataPreprocessingSteps.docx
- readme_RDANN.doc
Technical notes and papers
- RDANN_JAMC.docx
- McCandless+etal_2016_RE.pdf
- McCandless, T.C., Young, G.S., Haupt S.E., and L.M. Hinkelman, 2016: Regime Dependent Solar Irradiance Forecasting. Journal of Applied Meteorology and Climatology, Accepted in final form March 2016.