Fluid Mechanics and Acoustics Laboratory - UMR 5509

LMFA - UMR 5509
Laboratoire de Mécanique des Fluides et d’Acoustique
Lyon
France


Supervisory authorities

Our partners




Home > Pages perso > Vidali Cristina

Vidali Cristina

Vidali Cristina

PhD student - project CIFRE
LMFA and Air Liquide Paris Research & Development Center.
Subject: "Atmospheric heavy gas dispersion in complex environment"
Supervisor: Pietro Salizzoni & Louis Gostiaux

Understanding the physics of atmospheric dispersion of heavy gases plays a key role in evaluation and management of risks associated to accidental releases in complex environment.
Heavy gases dispersion is influenced by reduced dilution with ambient air and stagnation close to ground level with serious consequences in health and environmental impact. Their presence may produce favorable conditions to asphyxia, explosion and fire in urban areas or industrial sites.
The aim of this study is to investigate, experimentally and numerically, the dynamics of a heavy gas release trough measurements in air wind tunnel. In particular, we will identify local stratification influence on mixing between air and heavy gas, and on the multi-scale spectral distribution on turbulent kinetic energy. To that purpose, we defined the scenario of interest with the industrial partner Air Liquide as the emission from an Air Separation Units (ASU) that releases O₂ at a temperature of -40°C in the atmospheric boundary layer. We simulate it with a scale model in our wind tunnel facility, where the inflow condition has been set to reproduce a fully developed turbulent boundary layer over a rough surface in neutral condition. From an elevated source we release a dense mixture of carbon dioxide and ethane, the latter used as a tracer in concentration measurements. Under the same flow and emission set-up we reproduce a passive scalar release, employing a mixture of air and ethane, comparing the dataset with the heavy gas one. In the experiments, we perform simultaneous velocity and concentration measurement by means of a a coupled system, composed of a Flame Ionization Detector and a Hot-Wire Anemometry.
In the data analysis, focus is set on the mean concentration field and its higher-order moments, the concentration probability distribution and the turbulent mass fluxes, characterising the spectra, the turbulent kinetic energy exchange, the mixing on large and small scale, as well as the temporal structure of the signal.
Finally, the data collected during the experiments on the heavy gas release is used to test and validate two operational dispersion models. To that purpose we consider an integral model (Ventjet, Miller et al., 2021), developed by Air Product and Air Liquide, and a Lagrangian model (SLAM, Vendel et al., 2011), developed by the team AIR of the École Centrale de Lyon.
We complete the study by investigating the structure of the concentration time series, estimating the crossing time and rate of a concentration threshold by means of analytic models.