Model calibration is important as its functions improve building models. A calibrated model of the TxAIRE Houses #1 and #2 have been developed using the OpenStudio software. The model was developed and calibrated with data of the year 2014 but can be used predict electricity consumption of any year.

An actual meteorological year (AMY) weather file was also created for the houses which was used for calibration. The data used to develop this weather file were measured on-site which improved the accuracy of the model. It was found that an error difference of -1.0 % and -4.2 % occurred between the Texas Allergy, Indoor Environment and Energy (TxAIRE) weather file and weather file acquired from Weather Analytics and EnergyPlus websites respectively.

Simulation results from both houses were compared to the measured utility bill data for 2014. The energy variables simulated are total house energy consumption, outdoor unit energy consumption, air handler unit energy consumption and internal electric load energy consumption. Also, temperature in the attic and the temperature of the house (living room) were compared. Results of calibration for House #1 produced quality parameters normalized mean bias error (NMBE) of -0.14 and coefficient of variation of the mean bias error (CVRMSE) of 23.51. For House #2, an NMBE of 0.79 and CVRMSE of 7.59 were obtained. Both calibration values meet the accepted tolerance for calibration as set out by ASHRAE.

Date of publication

Spring 5-16-2016

Document Type




Persistent identifier


Committee members

Nelson Fumo, Frederica Brown, Althea Arnold


Master of Science in Mechanical Engineering