skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: On the significance of modeling internal surface convection in dynamic whole-building simulation programs

Conference ·
OSTI ID:20085680

The well-stirred assumption with surface-averaged convection coefficients is commonly employed by building thermal simulation programs to model heat transfer from the room air to internal surfaces (such as walls and windows). This approach assumes the room air to be at a uniform temperature (the influence of temperature stratification is ignored) and characterizes the heat transfer with surface-averaged convection coefficients. A model of the IEA Annex 21/Task 12 double-glazed test room was created and simulations were performed using the well-stirred modeling approach. Predictions of the room's heating energy consumption were found to be highly sensitive to both convection coefficients and to the temperature assumed for the room air. The sensitivity was so great that it is concluded that an accurate characterization of the convective regime is critical in order to accurately simulate configurations like the IEA test room. The simulation results were found to be far more sensitive to internal convection modeling than to other factors, such as the modeling of fabric thermal properties and air infiltration. In response to the demonstrated need for more accurate internal surface convection algorithms, recommendations are provided on an approach that will allow calculations to be responsive to local flow conditions.

Research Organization:
Natural Resources, Ottawa, Ontario (CA)
OSTI ID:
20085680
Resource Relation:
Conference: ASHRAE Annual Meeting, Seattle, WA (US), 06/18/1999--06/23/1999; Other Information: PBD: 1999; Related Information: In: ASHRAE Transactions: Technical and symposium papers presented at the 1999 annual meeting in Seattle, Washington of the American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc.; Volume 105, Part 2, by Geshwiler, M.; Harrell, D.; Roberson, T. [eds.], 1360 pages.
Country of Publication:
United States
Language:
English