# Simulation of Partially Premixed Combustion in a Co-axial Combustor

### Introduction

The purpose of this tutorial is to provide guidelines and recommendations for setting up and solving a reacting ﬂow using the partially premixed combustion model. This tutorial demonstrates how to do the following:

• Create a probability density function (PDF) ﬁle for a combustion system.
• Deﬁne ANSYS Fluent inputs for PDF chemistry modeling.
• Use the partially premixed Zimont turbulent ﬂame speed model to simulate the combustion system.
• Solve the case with appropriate solver settings.
• Postprocess the results to investigate the premixed and non-premixed properties.

### Problem Description

The coaxial combustor considered is shown in the following ﬁgure. A swirler at the center of the combustor introduces the lean methane/air mixture (equivalence ratio=0.8) of temperature 300 K with an axial velocity of 50 m/s and swirl velocity of 30 m/s. Pure air at an axial velocity of 10 m/s and temperature 650 K is introduced from the outer tube to stabilize the ﬂame. The major species involved in the combustion process are CH4, O2, CO2, CO, H2O, and N2.

The key to the premixed combustion model is the prediction of the turbulent ﬂame speed, the turbulent ﬂame speed normal to the mean surface of the ﬂame. The turbulent ﬂame speed is inﬂuenced by the following:

• Laminar ﬂame speed, which is, in turn, determined by the fuel concentration, temperature, and molecular diﬀusion properties, as well as the detailed chemical kinetics
• Flame front wrinkling and stretching by large eddies, and ﬂame thickening by small eddies

Numerous turbulent ﬂame speed models have been derived. You will be using the Zimont Turbulent Flame Speed Model (for details refer to section 9.4.1 of the ANSYS Fluent 14.5 Theory Guide).

### Result

The partially premixed model in ANSYS Fluent can be used to simulate a combustion system, where the combustion process is neither purely premixed nor purely non-premixed. Both premixed and non-premixed properties can be investigated using the postprocessing results.

### Summary

This tutorial demonstrated the application of the partially premixed model, based on both non-premixed (mixture-fraction based) and premixed (mean progress variable based) models.