Conference Year



building performance, retrofit, retrofit analysis, model validation, numerical support


It is well known that residential and commercial buildings account for about 40% of the overall energy consumed in the United States, and about the same percentage of CO2 emissions. Retrofitting existing old buildings, which account for 99% of the building stock, represents the best opportunity of achieving challenging energy and emission targets. United Technologies Research Center (UTC) has developed a methodology and tool that provides computational support for analysis and decision-making for building retrofits. The tool is based on simplified physics-based models and incorporates intelligent defaulting capability, automatic model calibration and package selection as well as uncertainty and sensitivity analysis on both predicted energy consumption and potential savings. The latter is used to better inform decision makers on the quality of the data used for analysis and direct them in the overall process to achieve the required accuracies in the analysis. This paper addresses the validation of the simplified physics-based models. The validation is performed using three-tier approach: a) validation against ASHRAE 140 BESTEST Cases; b) inter-model comparison of results obtained by other more complex tools using more detailed models than in those required by ASHRAE 140 Standard and c) comparison to real building measured utility data. Findings and conclusions from each one of the three validation approaches are presented, as well as a discussion on model complexity vs. results accuracy based on lessons learned during the reported study. This material is based upon work supported by the Energy Efficient Buildings Hub (EEB Hub), an energy innovation hub sponsored by the U.S. Department of Energy under Award Number DE-EE0004261 and by the U.S Department of Defense ESTCP Program ESTCP Program # 201257, contract number W912HQ-12-C-0051.

3263_presentation.pdf (442 kB)
Validation of retrofit analysis simulation tool: Lessons learned