Fire - Adaptive Mesh Refinement Solver for Compressible Reacting Flows

Fire is an open-source CFD solver specializing in compressible reacting flows, built upon the ECOGEN-v2.0 framework with enhanced combustion modeling capabilities. Key features include:

• 🚀 Advanced AMR strategies for flame and wave front resolution
• 🔥 Integrated Cantera chemistry solver
• 📐 Multi-dimensional simulation support (1D-3D)
• 🧪 Validated against benchmark combustion cases

Table of Contents

1. Installation
2. Example Cases
3. Post-Processing
4. Input File Configuration
5. Architecture Overview

---

1. Installation

Prerequisites

• Linux OS (Ubuntu 20.04/RedHat tested)
• MPI implementation (OpenMPI recommended)
• Conda package manager
• Cantera 3.0.0b1 with C++ interface

Step-by-Step Setup

1. Install libcantera-devel 3.0.0b1

conda create --name ct --channel cantera/label/dev libcantera-devel==3.0.0b1

2. Install MPI (Ubuntu)

sudo apt update && sudo apt install openmpi-bin openmpi-common libopenmpi-dev

3. Build Fire Solver

# From any example case directory:
./run.sh [options]

# Available options:
#   --debug    - Debug build
#   -j N       - Use N parallel processes
#   --no-run   - Build only

Note
The conda environment must be named ct for automatic build detection

---

2. Example Cases

Case Name	Description	Dimensions	Reference
1D_detonation	Gaseous detonation wave	1D	Paolucci 2014
1D_H2-Air_deflagration	Gaseous deflagaration wave	1D	None
2D_ISBI_Case_[1-5]	Inert shock-bubble interaction with different grid setups	2D	Ding 2018
2D_RSBI	Reactive shock-bubble interaction	2D	Diegelmann 2016
2D_Channel_detonation_[1-2]	Detonation with cellular structures using HLLC or HLLC-LM schemes	2D	Paolucci 2014
3D_Channel_detonation	Large-scale & challenging detonation simulation	3D	None

Case Setup Instructions:

• For each case, the setup files are located in the input_files/.
• Detailed instructions can be found in manuals of ECOGEN-v2.0.

Avoid spaces in paths

• Ensure all directory paths (including those for Fire or case files) contain no spaces.
  Valid example: /home/user1/test_case/Fire
  Invalid example: /home/user1/test case/Fire

Quick Start:

cd example_cases/1D_detonation
./run.sh

cd example_cases/2D_RSBI
./run.sh

---

3. Post-Processing

Simulation outputs are stored in results/[Case_Name]/ in the case folder with:

• AMR grid data in *.vtu format in datasets/ folder
• Time metadata in infoCalcul.out
• ParaView collection file collectionParaview.pvd

Recommended visualization workflow:

• Open collectionParaview.pvd in ParaView
• Use pMax field for cellular structure analysis
• 1D case results can be plotted using the plot_1D_*.py script prepared in the 1D case folders.

---

4. Input File Configuration

Essential Parameters

Parameter	Location	Values	Description
riemann_type	meshV5.xml	1(HLL), 2(HLLC), 3(HLLC-LM)	Inviscid flux scheme selection
gradRho_flag	meshV5.xml	0/1	Turn on/off the 3rd AMR strategy
lvlHydro/lvlChem	meshV5.xml	Integer	AMR level limits for hydrodynamical and chemical processes
CanteraInput	modelV4.xml	Path to YAML	Chemistry configuration

Critical Notes

• The CanteraInput in modelV4.xml is required even for inert simulations to model the thermodynamic and transport properties.
• The TOTAL_SPECIES_NUMBER in CMakeLists.txt shall be the same with the species number in the Cantera file.
• Transport models require additionalPhysic type="viscosity" in modelV4.xml

---

5. Architecture Overview

Class Inheritance

• White: ECOGEN base classes
• Red: Fire-specific implementations
• Green: Cantera integrations

Simulation Workflow

Fire-Specific Customizations (Red Highlights):

🔴 Overridden ECOGEN methods for:

• Hydrodynamic processes for the multi-component thermally perfect gas
• Chemical source term integration
• AMR refinement criteria

---

Acknowledgments
This work builds upon the ECOGEN framework. Please cite both projects when using Fire in academic publications.