Towards an integrated computational method to determine internal spaces for optimum environmental conditions

Tools

Sofotasiou, Polytimi and Calautit, John Kaiser and Hughes, Ben Richard and O'Connor, Dominic (2016) Towards an integrated computational method to determine internal spaces for optimum environmental conditions. Computers and Fluids, 127. pp. 146-160. ISSN 0045-7930 (https://doi.org/10.1016/j.compfluid.2015.12.015)

[thumbnail of Sofotasiou-etal-CF2016-Towards-integrated-computational-method-determine-internal-spaces-optimum-environmental-conditions]
Preview
 Text. Filename: Sofotasiou_etal_CF2016_Towards_integrated_computational_method_determine_internal_spaces_optimum_environmental_conditions.pdf
Accepted Author Manuscript
License: Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 logo
Download (1MB)| Preview

Abstract

Computational Fluid Dynamics tools and Response Surface Methodology optimization techniques were coupled for the evaluation of an optimum window opening design that improves the ventilation efficiency in a naturally-ventilated building. The multi-variable optimization problem was based on Design of Experiments analysis and the Central Composite Design method for the sampling process and estimation of quadratic models for the response variables. The Screening optimization method was used for the generation of the optimal design solution. The generated results indicated a good performance of the estimated response surface revealing the strength correlations between the parameters. Window width was found to have greater impact on the flow rate values with correlation coefficient of 73.62%, in comparison to the standard deviation 55.68%, where the window height prevails with correlation coefficient of 96.94% and 12.35% for the flow rate. The CFD results were validated against wind tunnel experiments and the optimization solution was verified with simulation runs, proving the accuracy of the methodology followed, which is applicable to numerous environmental design problems.

CORE (COnnecting REpositories)