Effects of manikin model simplification on CFD predictions of thermal flow field around human bodies

Simplified computational thermal manikins are beneficial to the computational efficiency of computational fluid dynamics simulations. However, the criterion of how to simplify a computational thermal manikin is still absent. In this study, three simplified computational thermal manikins (CTMs 2, 3 and 4) were rebuilt based on a detailed 3D scanned manikin (CTM 1) using different simplification approaches. Computational fluid dynamics computations of the human thermal plume in a quiescent indoor environment were conducted. The predicted airflow field using CTM 1 agreed well with the experimental observations from the literature. Although the simplified computational thermal manikins did not significantly affect the airflow predictions in the bulk regions, they strongly influenced the predicted airflow patterns near the computational thermal manikins. The predictive error of the computational thermal manikin was strongly related to the simplification approach. The computational thermal manikins generated from the surface-smoothing approach (CTM 2) was very close to CTM 1, while the required mesh elements for a stable numerical solution dropped by over 75%. Comparatively, the predictive errors of CTMs 3 and 4 were considerable in the near-body regions. This study has illustrated the importance of keeping the key body features when simplifying a computational thermal manikin. The surface-smoothing-based simplification method was shown to be a promising approach.