A numerical technique for mesh refinement in the HeaRT (Heat Release and Turbulence) numerical code is presented. In the CFD framework, Large Eddy Simulation (LES) approach is gaining in importance as a tool for simulating turbulent combustion processes, also if this approach has an high computational cost due to the complexity of the turbulent modeling and the high number of grid points necessary to obtain a good numerical solution. In particular, when a numerical simulation of a big domain is performed with a structured grid, the number of grid points can increase so much that the simulation becomes impossible: this problem can be overcomed with a mesh refinement technique. Mesh refinement technique developed for HeaRT numerical code (a staggered finite difference code) is based on an high order reconstruction of the variables at the grid interfaces by means of a moving least square interpolation: numerical code is written in modern Fortran (2003 standard of newer) and is parallelized using domain decomposition and message passing interface (MPI) standard. The multi-resolution algorithm is been validated by the comparison with DNS and LES data.

Multi-resolution techniques for a compressible staggered LES numerical code

ROSSI, GIACOMO
2015

Abstract

A numerical technique for mesh refinement in the HeaRT (Heat Release and Turbulence) numerical code is presented. In the CFD framework, Large Eddy Simulation (LES) approach is gaining in importance as a tool for simulating turbulent combustion processes, also if this approach has an high computational cost due to the complexity of the turbulent modeling and the high number of grid points necessary to obtain a good numerical solution. In particular, when a numerical simulation of a big domain is performed with a structured grid, the number of grid points can increase so much that the simulation becomes impossible: this problem can be overcomed with a mesh refinement technique. Mesh refinement technique developed for HeaRT numerical code (a staggered finite difference code) is based on an high order reconstruction of the variables at the grid interfaces by means of a moving least square interpolation: numerical code is written in modern Fortran (2003 standard of newer) and is parallelized using domain decomposition and message passing interface (MPI) standard. The multi-resolution algorithm is been validated by the comparison with DNS and LES data.
27-nov-2015
Inglese
parallel computing
FAVINI, Bernardo
GIACOMAZZI, Eugenio
GAUDENZI, Paolo
Università degli Studi di Roma "La Sapienza"
File in questo prodotto:
File Dimensione Formato  
PhD_Thesis.pdf

accesso aperto

Dimensione 28.57 MB
Formato Adobe PDF
28.57 MB Adobe PDF Visualizza/Apri

I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.14242/91458
Il codice NBN di questa tesi è URN:NBN:IT:UNIROMA1-91458