Abstract
whole-building energy simulation program that engineers, architects, and researchers use to model energy use in buildings. Calibration of a building model to utility data is often essential for the simulation engine to produce realistic estimates of energy use. These calibrated energy models are useful for commissioning of building systems, predictions of savings from applying energy conservation measures, and measurement and verification of building retrofit projects. A typical building has an average of around 3,000 parameters that could be calibrated. Manual calibration for only dozens of parameters could require months. Autotune is a novel methodology aimed at automatically producing calibrated building energy models using measured data such as utility bills, sub-meter data, and/or sensor data. The methodology creates models accurately and robustly by deriving near-optimal input parameter values in a systematic, automated, and repeatable fashion.
This paper independently demonstrates the application of Autotune in two case studies. In the first, a building model it is de-tuned by deliberately injecting faults into more than 60 parameters. This model was then calibrated using Autotune and its accuracy with respect to the original, manually-calibrated model was evaluated in terms of the industry-standard normalized mean bias error and coefficient of variation of root mean squared error metrics set forth in ASHRAE Guideline 14. In addition to whole-building energy consumption, output parameters including lighting, plug load profiles, HVAC energy consumption, zone temperatures, and other variables are analyzed. In the second, Autotune is compared directly to manual calibration of an emulated-occupancy, full-size residential building. The paper concludes with a discussion of the key strengths and weaknesses of auto-calibration approaches.
