Numerical study of air jet flow field during a loss of vacuum

• Numerical simulations of air jet flow field during a LOVA event.
• The SST model is the model that offers a realistic characterization of the turbulent flow compared to experimental data.
• The SST model is identified to be a model able to represent accurately the fluid dynamic field inside the VV with a simulation error of 13%.

Air leakage into tokamaks vacuum vessel during plasma burning or maintenance operations may lead to the fast pressurization of the vacuum vessel. A fraction of the dust inventory present in the vacuum vessel can be mobilized threatening the safety of staff and workers on site, the local population and the environment. A numerical analysis of the physical phenomena involved in such accidents is necessary in order to predict the thermal-fluid dynamics into the vacuum vessel after air ingress and consequent dust mobilization. Accuracy of the numerical results is also required in order to provide a sufficient margin in the design of the safety systems. The numerical simulation of Loss of Vacuum Accident (LOVA) scenarios is a challenging task for today numerical methods and models because it involves large volumes, multiphase flows ranging from highly supersonic to nearly incompressible and contemporary heat transfer. The drag force exerted on the dust by a moving fluid due to the viscous surface shear stress and pressure distribution around the dust particles depends mainly on the Reynolds number, i.e. property of the fluid (kinematic viscosity), its mean velocity and characteristic length of the geometry. For a fixed geometry, the key parameter for the dust mobilization is the velocity field of the continuous phase, and its thermodynamics properties, inside the vacuum vessel. In this contribution, the authors present and discuss the results of numerical simulations of air jet flow field during a LOVA with particular attention to the comparison with the experimental data and differences arising from the use of different types of grid resolution and turbulence models (Zero-Equation, k–ω and SST).