A three-dimensional thermoregulatory model for predicting human thermophysiological responses in various thermal environments

This study presents a complete virtual three-dimensional (3D) model of human thermophysiological responses by integrating detailed heat dissipation model from human skin to environment through spatially heterogeneous (realistic) clothing into human thermoregulation model by two-way coupling. The model considered various heat and mass transfer mechanisms in air and clothing layers such as conduction, natural convection, forced convection, radiation, and evaporation. The numerical setup of the model was presented and simulated using ANSYS Fluent. The developed model is systematically validated against experimental data obtained from authors of published literatures. To demonstrate the robustness and accuracy of the model, it has been validated for a wide range of ambient temperatures (−10.0 to 45 °C), relative humidity (25–86%), activity levels (0.8–3.5 mets), clothing thermal resistances (0.10–2.95 clo), and evaporative resistances (11.5–800 Pa m 2 /W) including steady-state and transient exposures. This complete 3D model predicted human thermophysiological responses for a wide range of ambient condition with good accuracy. The average RMSD (root mean square deviation) of predicted core and mean skin temperatures was 0.21 °C (maximum RMSD = 0.33 °C) and 0.56 °C (maximum RMSD = 1.03 °C), compared to average standard deviation in measured data 0.26 °C and 0.58 °C, respectively. The model is also able to predict the local skin temperature with an average RMSD of 0.64 °C. Therefore, this newly developed model can be served as a useful tool in analysing human thermophysiological responses and heat stress for any given thermal environment and clothing setting. • A complete virtual 3D thermoregulation model was developed using CFD. • Average RMSD of predicted core and mean skin temperatures was 0.21 °C and 0.56 °C. • The established model is a useful tool for analysing human thermal comfort. • Model is validated against data obtained from authors of published literatures.