Analysis of vortex core generation in pipe flows under different reynolds number conditions

[EN] Pipe flow is a well-documented case widely studied in both theoretical and practical applications. The present work aims at studying the influence of the Reynolds number on turbulent vortex distribution using Large Eddy Simulations (LES). Features such as the mean velocity profiles and root mean squared velocity are first numerically investigated for different fluid properties involving Reynolds numbers ranging from 5,925 to 15,190 in order to verify the law-of-the-wall and turbulence statistics with experimental and DNS data. Once the simulations are validated, the vortex core generation within the flow is studied through a detection algorithm based on the lambda 2 criterion with two different approaches, first using an absolute threshold value and then using a relative threshold value depending on the turbulent intensity. Results are compared in terms of number of structures and Probability Density Functions for both the size and the radial distributions. Finally, results are compared for one condition with the Q-criterion to assess the results obtained resulting in practically identical volume and radial distributions. These results are deemed to shed light on the vortex formation and location to generate proper inflow boundary conditions to highly resolved simulations in varied engineering applications.
This research has been funded by the Spanish Ministerio de Economia y Competitividad through the project RTI2018099706-B-100: "Estudio de la atomizacion primaria mediante simulaciones DNS y tecnicas opticas de muy alta resolucion" and the Spanish Ministerio de Ciencia e innovacion through the project EQC2018004605-P: "Estudio del proceso de inyeccion en atmosferas presurizadas". The authors thankfully acknowledge the computer resources from the Rigel cluster at UPV (Spain) and the Bebop cluster from the Laboratory Computing Resource Center at Argonne National Laboratory (USA). L.A. Gonzalez-Montero is partially supported through the contract FPI -Subprograma 2 of the Universitat Politecnica de Valencia.