Investigation of inhalation and exhalation flow pattern in a realistic human upper airway model by PIV experiments and CFD simulations.

In this study, flow field characteristics in the trachea region in a realistic human upper airway model were firstly measured by particle image velocimetry (PIV) in the air under three constant inhalation and exhalation conditions: 36 L/min, 64 L/min and 90 L/min, representing flow rates of 18 L/min, 32 L/min and 45 L/min in real human airway (the model was twice the size of a human airway). Computational fluid dynamics (CFD) analyses were performed on four turbulence models, with boundary conditions corresponding to the PIV experiments. The effects of flow rates and breathing modes on the airflow patterns were investigated. The CFD prediction results were compared with the PIV measurements and showed relatively good agreement in all cases. During inhalation, the higher the flow rates, the less significant the "glottal jet" phenomenon, and the smaller the area of the separation zone. The air in the nasal inhalation condition accelerated more dramatically after glottis. The SST-Transition model was the best choice for predicting inhalation velocity profiles. For exhalation condition, the maximum velocity was much smaller than that during inhalation due to the more uniform flow field. The exhalation flow rates and breathing modes had little effect on the flow characteristics in the trachea region. The RNG k - ε model and SST k - ω model were recommended to simulate the flow field in the respiratory tract during exhalation.