Calibrated dynamic zonal model DOMA+ using the SCE-UA method – application to atrium temperature distribution prediction

Abstract

Atriums in modern building designs often represent particular problems related to energy and airflow distribution, such as thermal stratification, local overheating, and uncontrolled contaminant spreading. This study developed a new dynamic zonal model with air-diffuser (DOMA) to predict the airflow and temperature stratification of an atrium, and then it was experimentally calibrated (known as DOMA+) by using the Shuffled Complex Evolution method developed at the University of Arizona (SCE-UA). Two-parameter set groups were considered during the calibration, which are Group One, including the flow coefficient K and the convective heat transfer coefficients for surfaces of the atrium; in Group Two, the flow coefficient K is separated into K-horizontal and K-vertical representing the flow coefficient used in horizontal and vertical directions. The results show that although the Group-One calibrated zonal model with one K is accurate to predict space thermal behavior, an update to two Ks (Group Two) is suggested to improve the zonal model prediction accuracy and quality. The use of the SCE-UA model reveals the inherent limitations and characteristics of the zonal model DOMA and thus indicates how to improve the prediction ability of a zonal modeling approach to advance its development.