Statistical Analysis of Neural Networks as Applied to Building Energy Prediction

[+] Author and Article Information
Robert H. Dodier

60 South Boulder Circle #6301, Boulder, CO 80303 USA

Gregor P. Henze

Architectural Engineering, University of Nebraska–Lincoln, Peter Kiewit Institute, 1110 South 67th Street, Omaha, Nebraska 68182-0681 USA

J. Sol. Energy Eng 126(1), 592-600 (Feb 12, 2004) (9 pages) doi:10.1115/1.1637640 History: Received April 01, 2002; Revised May 01, 2003; Online February 12, 2004
Copyright © 2004 by ASME
Your Session has timed out. Please sign back in to continue.


Haberl,  J., and Thamilseran,  S., 1996, “Predicting Hourly Building Energy Use: The Great Energy Predictor Shootout II: Measuring Retrofit Savings,” ASHRAE Trans., 102, Part 2.
MacKay,  D., 1994, “Bayesian Nonlinear Modeling for the Prediction Competition,” ASHRAE Trans., 100(2), pp. 1053–1062.
Ohlsson,  M., Peterson,  C., Pi,  H., Rognvaldsson,  T., and Soderberg,  B., 1994, “Predicting System Loads with Artificial Neural Networks,” ASHRAE Trans., 100(2), pp. 1063–1074.
Feuston,  B., and Thurtell,  J., 1994, “Generalized Nonlinear Regression with Ensemble of Neural Nets,” ASHRAE Trans., 100(2), pp. 1075–1080.
Cybenko,  G., 1989, “Approximation by Superpositions of a Sigmoidal Function,” Math. Control, Signals, Syst., 2, 303–314.
Hornik,  K., Stinchcombe,  M., and White,  H., 1989, “Multilayer Feedforward Networks are Universal Approximators,” Neural Networks, 2, pp. 359–366.
Dodier, R., 1995, “Statistical Properties of Neural Networks, with Application to Building Energy Prediction,” Master’s thesis. U. Colorado, Boulder, Colorado.
Kreider,  J., Claridge,  D., Curtiss,  P., Dodier,  R., Haberl,  J., and Krarti,  M., 1995, “Building Energy Use Prediction and System Identification using Recurrent Neural Networks,” J. Sol. Energy Eng., 117(3), pp. 161–166.
Seber, G. A. F., and Wild, C. J. 1989, Nonlinear Regression, John Wiley and Sons, New York.
Raftery, A., 1994, Bayesian Model Selection in Social Research, in Sociological Methodology 1995, edited by P. Marsden, Cambridge, Mass., Blackwells.
Katipamula,  S., 1996, “The Great Energy Predictor Shootout II: Modeling Energy Use in Large Commercial Buildings,” ASHRAE Trans., 100 (2).
Abarbanel,  H. D. I., Brown,  R., Sidorowich,  J., and Tsimring,  L. S., 1993, “Analysis of Observed Chaotic Data in Physical Systems,” Rev. Mod. Phys., 65(4), pp. 1331–1392.
Press, W., Flannery, B., Teukolsky, S., and Vetterling, W., 1988, Numerical Recipes in C, Cambridge, Cambridge University Press.
Raftery,  A., Madigan,  D., and Hoeting,  J. A., 1997, “Bayesian Model Averaging for Linear Regression Models,” J American Statistical Assoc, 437, pp. 179–191.


Grahic Jump Location
A χ2-distribution with the mean and critical values of W indicated. The tail mass shown here is α=0.001.
Grahic Jump Location
Short-term normalized autocovariance functions, at lags from 1 to 24 hours. At left, from the top downwards, there are temperature, insolation, and wind.
Grahic Jump Location
Long-term normalized autocovariance function of ambient temperature, at lags from 1 hour to 50 days
Grahic Jump Location
Target values (solid line) of CHW, Business Building, compared to predictions (dotted line) made by a network. When a prediction is too high or too low, the next prediction tends to be too high or too low also.




Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In