Flow and contaminant transfer at the interface of an aerodynamic containment barrier subjected to the wake of a moving obstacle

This contribution is an experimental and numerical study on the breach of a containment barrier subject to the transient wake of a moving obstacle. In the context of a 100 m3 laboratory with high air exchange rate (8.5 h-1), a passive gaseous contaminant is emitted at a controlled flow rate from the bottom surface of an open cavity (0.8 x0.34 m2 wide) swept by a continuous airflow drawn from the lab. Starting from a steady and turbulent airflow situation, a human-sized obstacle passes in front of the cavity and disturbs the established flow. The containment breach is then analyzed by recording time-dependent gas concentrations and velocity components, measured at the cavity open surface, with photoionization detectors and 2D3C PIV, respectively. A penalty method coupled with LES is used to model the experimental setup and predictions are compared to measurements. The numerical model is found to be able to reproduce the contaminant concentrations, air velocities and main POD modes of the flow within experimental inaccuracies. This modeling approach appears to be suitable to study, in-silico, the effect of drafts induced by human motion on aerodynamic barriers. CFD results also show that pollutant leakage is inherently variable, even under stable ventilation conditions. Up to 100% variation of contaminant concentration peaks can be observed, depending on the initial turbulent state of the room.

Automated summary

This contribution presents an experimental and numerical study on the breach of a containment barrier caused by the transient wake of a moving obstacle. The study shows that pollutant leakage is inherently variable, even under stable ventilation conditions, and up to 100% variation in contaminant concentration peaks can be observed depending on the initial turbulent state of the room. The numerical model used in this study is found to be able to accurately reproduce the experimental results within the experimental inaccuracies.