Many building energy prediction models have been developed during the past decades. While popular tools such as Energy Star target single-use buildings, few have focused on mixed-use buildings due to its complexity. In practice, most non-residential buildings are mixed-use buildings supporting various functions such as office, cafeteria, public area, etc. The prediction models developed by Energy Star are based on the building categories defined in Commercial Building Energy Consumption Survey (CBECS), which consider only primary building activities instead of all activity types. This limitation compromises the model’s accuracy. This paper aims to tackle this challenge of energy prediction in mixed-use buildings. By applying simulation and statistical techniques, the proposed method reflects on both modeling and empirical approaches to diminish the difficulty in predicting mixed-use buildings energy consumption. Validation of this method was conducted by comparing the predicted enrgy use to the actual energy consumption data. It was shown that the proposed method was effective for energy prediction in mixed-use buildings. In addition, this approach inherits the complexity and yet efficiency from both simulation and statistical approaches.
[1]
Roberto Lamberts,et al.
BUILDING SIMULATION CALIBRATION USING SENSITIVITY ANALYSIS
,
2005
.
[2]
Karline Soetaert,et al.
xsample(): An R Function for Sampling Linear Inverse Problems
,
2009
.
[3]
Nance E. Matson,et al.
Review of California and National Methods for Energy PerformanceBenchmarking of Commercial Buildings
,
2005
.
[4]
Mary Ann Piette,et al.
Development of a California commercial building benchmarking database
,
2002
.
[5]
R. Sharp,et al.
Benchmarking Energy Use in Schools
,
1998
.
[6]
David E. Claridge,et al.
Calibration Procedure for Energy Performance Simulation of a Commercial Building
,
2003
.
[7]
Aldomar Pedrini,et al.
A methodology for building energy modelling and calibration in warm climates
,
2002
.