Performance Evaluation of RANS-Based Turbulence Models in Predicting Turbulent Non-Premixed Swirling Combustion within a Realistic Can Combustor

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kumar, Aishvarya, Bharti, Ram Prakash
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918511731802112
author Kumar, Aishvarya
Bharti, Ram Prakash
author_facet Kumar, Aishvarya
Bharti, Ram Prakash
contents This study has presented a comprehensive computational fluid dynamics (CFD) analysis of combustion flow in a realistic can combustor, evaluating the influence of various turbulence models on flow, thermal, and species fields. The non-premixed combustion modeling is performed using a presumed (beta) PDF approach in conjunction with a steady laminar flamelet model employing the San Diego reaction mechanism, and the turbulence is modeled using the RANS approach. The influence of turbulence models (standard $k-ε$, realizable $k-ε$, SST $k-ω$, LPS-RSM) on the velocity field, such as the mean axial velocity, mean transverse velocity, turbulent kinetic energy (TKE) and shear stress, is analyzed, besides their influence on temperature and species (\ce{C3H8}, \ce{CO2}, and \ce{CO}) concentration. Analysis showed that despite the shortcomings of the isotropic turbulent viscosity formulation of the SST $k-ω$ model being evident, it predicted the mean axial velocity, mean transverse velocity, turbulent kinetic energy and shear stress more accurately. Additionally, it predicted the flow features expected in a can combustor, such as the central recirculation zone (CRZ) and central vortex core (CVC), more accurately than other models. Besides, the model predicted a higher temperature in the primary zone, which is supported by a lower prediction of \ce{C3H8}, and elevated TKE, both of which support strong mixing and efficient heat release. Furthermore, the SST $k-ω$ model predicted the most compact stoichiometric mixture fraction bubble, encompassing CRZ and shear layers, indicating that the majority of the combustion occurs in the primary zone. The corresponding progress variables also indicated high values in the primary zone and shear layers, confirming near completion of the reaction, supported by negligible prediction of \ce{C3H8} and \ce{CO} at the outlet.
format Preprint
id arxiv_https___arxiv_org_abs_2605_19994
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Performance Evaluation of RANS-Based Turbulence Models in Predicting Turbulent Non-Premixed Swirling Combustion within a Realistic Can Combustor
Kumar, Aishvarya
Bharti, Ram Prakash
Fluid Dynamics
This study has presented a comprehensive computational fluid dynamics (CFD) analysis of combustion flow in a realistic can combustor, evaluating the influence of various turbulence models on flow, thermal, and species fields. The non-premixed combustion modeling is performed using a presumed (beta) PDF approach in conjunction with a steady laminar flamelet model employing the San Diego reaction mechanism, and the turbulence is modeled using the RANS approach. The influence of turbulence models (standard $k-ε$, realizable $k-ε$, SST $k-ω$, LPS-RSM) on the velocity field, such as the mean axial velocity, mean transverse velocity, turbulent kinetic energy (TKE) and shear stress, is analyzed, besides their influence on temperature and species (\ce{C3H8}, \ce{CO2}, and \ce{CO}) concentration. Analysis showed that despite the shortcomings of the isotropic turbulent viscosity formulation of the SST $k-ω$ model being evident, it predicted the mean axial velocity, mean transverse velocity, turbulent kinetic energy and shear stress more accurately. Additionally, it predicted the flow features expected in a can combustor, such as the central recirculation zone (CRZ) and central vortex core (CVC), more accurately than other models. Besides, the model predicted a higher temperature in the primary zone, which is supported by a lower prediction of \ce{C3H8}, and elevated TKE, both of which support strong mixing and efficient heat release. Furthermore, the SST $k-ω$ model predicted the most compact stoichiometric mixture fraction bubble, encompassing CRZ and shear layers, indicating that the majority of the combustion occurs in the primary zone. The corresponding progress variables also indicated high values in the primary zone and shear layers, confirming near completion of the reaction, supported by negligible prediction of \ce{C3H8} and \ce{CO} at the outlet.
title Performance Evaluation of RANS-Based Turbulence Models in Predicting Turbulent Non-Premixed Swirling Combustion within a Realistic Can Combustor
topic Fluid Dynamics
url https://arxiv.org/abs/2605.19994