Wavelets, Self-organizing Maps and Artificial Neural Nets for Predicting Energy Use and Estimating Uncertainties in Energy Savings in Commercial Buildings

Wavelets, Self-organizing Maps and Artificial Neural Nets for Predicting Energy Use and Estimating Uncertainties in Energy Savings in Commercial Buildings. (August 2009) Yafeng Lei, B.S., Tianjin University; M.S., Texas A&M University Chair of Advisory Committee: Dr. Kris Subbarao This dissertation develops a “neighborhood” based neural network model utilizing wavelet analysis and Self-organizing Map (SOM) to predict building baseline energy use. Wavelet analysis was used for feature extraction of the daily weather profiles. The resulting few significant wavelet coefficients represent not only average but also variation of the weather components. A SOM is used for clustering and projecting high-dimensional data into usually a one or two dimensional map to reveal the data structure which is not clear by visual inspection. In this study, neighborhoods that contain days with similar meteorological conditions are classified by a SOM using significant wavelet coefficients; a baseline model is then developed for each neighborhood. In each neighborhood, modeling is more robust without unnecessary compromises that occur in global predictor regression models. This method was applied to the Energy Predictor Shootout II dataset and compared with the winning entries for hourly energy use predictions. A comparison

[1]  David E. Claridge,et al.  Uncertainty of “Measured” Energy Savings from Statistical Baseline Models , 2000 .

[2]  R. Norgaard,et al.  Efficiency and sustainability in restructured electricity markets: the renewables portfolio standard , 1996 .

[3]  T. Agami Reddy,et al.  Prediction uncertainty of linear building energy use models with autocorrelated residuals , 1999 .

[4]  R. P. Mazzucchi,et al.  The project on restaurant energy performance end-use monitoring and analysis , 1985 .

[5]  David J. C. MacKay,et al.  BAYESIAN NON-LINEAR MODELING FOR THE PREDICTION COMPETITION , 1996 .

[6]  Saifur Rahman,et al.  Short-term load forecasting with local ANN predictors , 1999 .

[7]  T. A. Reddy,et al.  Uncertainty in baseline regression modeling and in determination of retrofit savings , 1998 .

[8]  Jeff Haberl,et al.  A Bin Method for Calculating Energy Conservation Retrofit Savings in Commercial Buildings , 1994 .

[9]  David E. Claridge,et al.  Ambient-temperature regression analysis for estimating retrofit savings in commercial buildings , 1998 .

[10]  M. Daneshdoost,et al.  Neural network with fuzzy set-based classification for short-term load forecasting , 1998 .

[11]  David E. Claridge,et al.  Multivariate Regression Modeling , 1998 .

[12]  Bryan Usevitch,et al.  A tutorial on modern lossy wavelet image compression: foundations of JPEG 2000 , 2001, IEEE Signal Process. Mag..

[13]  Martin T. Hagan,et al.  Neural network design , 1995 .

[14]  David West,et al.  A comparison of SOM neural network and hierarchical clustering methods , 1996 .

[15]  J. Stoer,et al.  Introduction to Numerical Analysis , 2002 .

[16]  Margaret F. Fels.  PRISM: An Introduction , 1986 .

[17]  Amitava Dhar.  Development of Fourier Series and Artificial Neural Network Approaches to Model Hourly Energy Use in Commercial Buildings , 1995 .

[18]  Josep Vehí,et al.  A hybrid system combining Self Organizing Maps with Case Based Reasoning in Structural Assessment , 2005, CCIA.

[19]  C. Brislawn.  Fingerprints Go Digital , 1997 .

[20]  Steven B. Smith,et al.  Digital Signal Processing: A Practical Guide for Engineers and Scientists , 2002 .

[21]  J. Neter,et al.  Applied Linear Regression Models , 1983 .

[22]  David E. Claridge,et al.  Baseline calculations for measurement and verification of energy and demand savings in a revolving loan program in Texas , 1998 .

[23]  David E. Claridge,et al.  ASHRAE's Guideline 14-2002 for Measurement of Energy and Demand Savings: How to Determine What Was Really Saved by the Retrofit , 2005 .

[24]  David E. Claridge,et al.  Predicting Energy Usage in a Supermarket , 1989 .

[25]  David E. Claridge,et al.  An expert system for building energy consumption analysis: prototype results , 1987 .

[26]  David E. Claridge,et al.  A Four-Parameter Change-Point Model for Predicting Energy Consumption in Commercial Buildings , 1992 .

[27]  Qing Yang,et al.  Modeling the effects of meteorology on ozone in Houston using cluster analysis and generalized additive models , 1998 .

[28]  A. Haar.  Zur Theorie der orthogonalen Funktionensysteme , 1910 .

[29]  J. K. Kissock.  A Methodology to Measure Retrofit Energy Savings in Commercial Buildings , 2008 .

[30]  Tomonobu Senjyu,et al.  A neural network based several-hour-ahead electric load forecasting using similar days approach , 2006 .

[31]  T. A. Reddy,et al.  A procedure to group residential air-conditioning load profiles during the hottest days in summer , 1990 .

[32]  Jerry M. Davis,et al.  An Automated Classification Scheme Designed to Better Elucidate the Dependence of Ozone on Meteorology , 1994 .

[33]  Hsin-Chung Lu,et al.  Prediction of daily maximum ozone concentrations from meteorological conditions using a two-stage neural network , 2006 .

[34]  Erkki Oja,et al.  Engineering applications of the self-organizing map , 1996, Proc. IEEE.

[35]  J. S. Haberl,et al.  Use of metered data analysis to improve building operation and maintenance: Early results from two federal complexes , 1988 .

[36]  Roy D. Wallen,et al.  The Illustrated Wavelet Transform Handbook , 2004 .

[37]  David E. Claridge,et al.  Baselining methodology for facility-level monthly energy use. Part 1: Theoretical aspects , 1997 .

[38]  Esa Alhoniemi,et al.  Self-organizing map in Matlab: the SOM Toolbox , 1999 .

[39]  S. Mallat,et al.  Multiresolution representations and wavelets , 1988 .

[40]  E. Vieth.  Fitting piecewise linear regression functions to biological responses. , 1989, Journal of applied physiology.

[41]  Moncef Krarti,et al.  Building Energy Use Prediction and System Identification Using Recurrent Neural Networks , 1995 .

[42]  Ari Rabl.  Parameter estimation in buildings: Methods for dynamic analysis of measured energy use , 1988 .

[43]  Jeff Haberl,et al.  The great energy predictor shootout. II : Measuring retrofit savings , 1998 .

[44]  David E. Claridge,et al.  EMODEL: A NEW TOOL FOR ANALYZING BUILDING ENERGY USE DATA , 1993 .

[45]  N. P. Leslie,et al.  Regression based process energy analysis system , 1986 .

[46]  Teuvo Kohonen,et al.  Self-Organizing Maps , 2010 .

[47]  Donald L. Hadley,et al.  Daily variations in HVAC system electrical energy consumption in response to different weather conditions , 1993 .

[48]  W. Pedrycz,et al.  Artificial neural network for forecasting daily loads of a Canadian electric utility , 1993, [1993] Proceedings of the Second International Forum on Applications of Neural Networks to Power Systems.

[49]  David E. Claridge,et al.  Inverse Model Toolkit: Application and Testing , 2003 .

[50]  David E. Claridge,et al.  Using synthetic data to evaluate multiple regression and principal component analyses for statistical modeling of daily building energy consumption , 1994 .

[51]  S. Vemuri,et al.  Neural network based short term load forecasting , 1993 .

[52]  Refrigerating.  2001 ASHRAE handbook : fundamentals , 2001 .

[53]  A. Grossmann,et al.  DECOMPOSITION OF HARDY FUNCTIONS INTO SQUARE INTEGRABLE WAVELETS OF CONSTANT SHAPE , 1984 .

[54]  Miriam L. Goldberg,et al.  Using the scorekeeping approach to monitor aggregate energy conservation , 1986 .

[55]  Joseph H. Eto,et al.  On using degree-days to account for the effects of weather on annual energy use in office buildings , 1988 .

[56]  Robert C. Sonderegger,et al.  A Baseline Model for Utility Bill Analysis Using Both Weather and Non-Weather Related Variables 1 , 1998 .

[57]  Sabaratnam Thamilseran.  An Inverse BIN Methodology to Measure the Savings from Energy Conservation Retrofits in Commercial Buildings , 1999 .

[58]  Gregor P. Henze,et al.  Statistical Analysis of Neural Networks as Applied to Building Energy Prediction , 2004 .

[59]  David E. Claridge,et al.  Inverse Modeling Toolkit: Numerical Algorithms , 2004 .

[60]  David E. Claridge,et al.  A Change-Point Principal Component Analysis (CP/PCA) Method for Predicting Energy Usage in Commercial Buildings: The PCA Model , 1993 .

[61]  Jeff Haberl,et al.  Predicting hourly building energy usage , 1994 .

[62]  J. M. MacDonald,et al.  Investigation of metered data analysis methods for commercial and related buildings , 1989 .

[63]  T. A. Reddy.  Application of Dynamic Building Inverse Models to Three Occupied Residences Monitored Non-Intrusively , .

[64]  Teuvo Kohonen,et al.  Self-organized formation of topologically correct feature maps , 2004, Biological Cybernetics.