Hydrogen instead of natural gas: optimal burner geometry thanks to CFD simulation
Why do I need to simulate combustion?
Hydrogen has different physical and chemical properties to natural gas. As a result, hydrogen combustion also behaves differently from that of natural gas. For example, hydrogen has a higher flame speed, lower ignition temperature and higher reactivity. This has an impact on flame stability, efficiency and emissions. As a result, if you simply use the same burner as the one designed for natural gas, you run the risk of not achieving the desired performance, and even of having safety problems.
For example, there may be flashback, self-ignition or incomplete combustion. This can lead to increased hydrogen consumption, reduced heat transfer or increased formation of nitrogen oxides (NOx). To avoid this, it is advisable to adapt the burner to hydrogen. This involves modifying the fuel-air ratio, pressure, temperature or even burner geometry. Perhaps you’d also like to analyze the behavior of different mixing ratios between natural gas and hydrogen? But how do you know which parameters to modify?
The answer is not so simple, because hydrogen combustion is a complex process that depends on many factors. Thanks to a powerful simulation tool, you can virtually model the flow, combustion and heat transfer in your burner and test different scenarios. This enables you to find the optimum configuration for your burner before using it in the real world.
Hydrogen combustion has special characteristics that must be taken into account when switching from natural gas to hydrogen | © CADFEM Germany
How can CFD simulation help me?
This is where Ansys Fluent comes in for CFD simulation. CFD stands for Computational Fluid Dynamics, and is a method of calculating the flow of gases and liquids numerically. CFD simulations allow you to virtually simulate your burner and vary various parameters. You can see how flow, combustion and heat transfer behave in your burner, and what the effects are on performance and emissions. You can also compare different burner designs and choose the best one.
Other benefits of CFD simulation for industrial burners include:
- Saves time and money: No need for actual testing with expensive prototypes.
- Increased safety: Anticipate and avoid flashbacks and spontaneous combustion.
- Improved efficiency and environmental compatibility: optimized fuel/air ratio and reduced NOx emissions.
To create the virtual prototype, define the geometry, create the calculation grid and select the combustion model | © CADFEM Germany
Simulating combustion – Starting with the mesh
Once you’ve mastered the first steps and clearly defined your task, it’s time to focus on mesh generation. Meshing a grid composed of individual cells has a significant influence on the accuracy and stability of the simulation. A high-quality mesh should always have the following properties:
- Sufficient resolution to capture all relevant flow structures,
- a smooth transition between regions to reduce numerical error, and
- alignment of cell surfaces as orthogonal as possible to current lines.
So what does a suitable mesh for hydrogen combustion look like? Here are a few tips on how to achieve the optimum mesh size [1]. The table below gives an orientation of the important mesh parameters for the burner itself, including the nozzles for the reactant feed, for the combustion chamber and its core, and for the outer zones where no reaction takes place. In detail, every CFD simulation always requires a sensitivity study to minimize the influence of mesh resolution. If turbulent flow structures and the interaction of chemistry and flow are to be calculated in particular detail, a much finer mesh may also be required.
Moreover, Ansys Fluent’s patented MOSAIC Poly Hexcore technology enables high-quality structured meshing in the combustion chamber, which can also be calculated excellently on a cluster. And with the Fault-Tolerant Meshing Workflow, you can also mesh difficult and complicated CAD geometries, as the method provides automatic correction mechanisms. If you’re not sure of the mesh resolution required, or simply want to resolve the flame front more accurately, you can use the adaptive mesh for polyhedra (PUMA algorithm). A predefined criterion based on the so-called flame indicator refines the mesh only when necessary [2].
| 1. Burner and combustion chamber | A body of influence (BOI) in the burner refines the mesh so that there are at least 20 to 25 cells in the diameter. It is preferable to extend the BOI into the combustion chamber by 3 to 4 times the burner diameter. |
| 2. Combustion chamber core | The BOI ensures uniform, high-quality meshing in the core, improving resolution of mixing, flame fronts and emissions. |
| 3. Burner inlet nozzles | 8 to 10 cells in the diameter of the hole are enough to predict whether the flame can be extinguished. |
| 4. External zones | The resolution outside the core can be coarser, ideally twice that of the core. |
To create a hydrogen burner, the combustion chamber can be divided into several zones.
With Ansys Fluent’s Poly-Hexcore mesh, you can create the appropriate mesh for your burner geometry. The different zones within (1-4) are explained in the table | © Ansys, CADFEM Germany GmbH
Which is the right combustion model? EDC vs. FGM model
We now turn to the selection of the appropriate combustion model. It must be chosen in such a way as to adequately and accurately describe the chemical reactions and the turbulence-chemistry interactions. The most common models are the EDC (Eddy Dissipation Concept) and the FGM (Flamelet Generated Manifold). The EDC model is based on the idea that chemical reactions occur in the smallest turbulent eddies. The reaction rate is influenced by the size and frequency of the eddies, which in turn depend on the turbulence. More specifically, the EDC model uses an extended reaction mechanism and solves the reaction equations using efficient tabulation algorithms such as the ISAT (In-Situ Adaptive Tabulation Algorithm).
On the other hand, the FGM model is based on the assumption that the internal flame structure follows a one-dimensional representation. This is calculated and tabulated in advance using a 1D laminar flame solver such as Chemkin. The FGM model can, for example, be combined with a skeletal reaction mechanism, which contains 10 gaseous components (species) and 25 reactions. If the number of reactions is not sufficient, the UCSD mechanism (57 species, 268 reactions) is an alternative. The table below compares the EDC and FGM models. The table below provides an example of FGM model setup for a partially premixed flame, as well as recommendations for numerical settings. Sufficiently long flow times ensure significant transient averaging of the results.
When we talk about combustion, we must also consider the turbulence model. In cases where flow fields are stable, RANS models (k-ω SST, k-ε realizable) provide an efficient solution for many engineering problems. However, there is often a strong coupling between the flow and the reaction. For example, unstable flows lead to local flame extinction. In order to capture the flame formation and pollutant emissions as accurately as possible, Large Eddy Simulation (LES) is the best choice. However, LES is generally considered very complex. A good compromise between accuracy and effort is Stress-Blended Eddy Simulation (SBES), because it can transition from RANS models to LES models.
A comparison of the EDC and FGM combustion models
| Combustion model | Advantages | Disadvantages |
| Eddy Dissipation Concept (EDC) | Ideal for complex chemistry or multiple fuels, as it uses detailed reaction mechanisms. | Requires more computational time because it needs to solve chemical equilibria continuously. Does not account for the flame structure and may therefore struggle to capture flame propagation or extinction. |
| Flamelet Generated Manifold (FGM) | Advantageous for applications with rapid reactions or partial premixing, such as gas turbines or processing furnaces, due to the low computational time. | May struggle to capture flashback or detonations. Calculating the flame structure requires more memory. |
Example setup: Hydrogen combustion with the FGM model for a partially premixed flame:
| Turbulence | Combustion | H2 Reaction Mechanism | Flow time (transient evaluation) |
| LES (Dynamic Smagorinsky), SBES | MGF & Finite Rate Model | Skeleton | 3-4 (Initialization), 7-8 (Statistics) |
| Discretization | Pressure-velocity coupling | No time step and sub-iterations |
| Momentum: Bounded Central Differencing, Temps : Bounded 2nd order implicit Autres: 2nd-order upwind | SIMPLEC | 1e-5s, 10 It. |
And now? Simulate with the GPU solver in Ansys Fluent.
Detailed and complex methods such as EDC and LES require significant computational power. But what if you don’t have access to a high-performance cluster? The new GPU solver in Ansys Fluent provides the solution, as the computation can be fully performed on one or more graphics cards. This allows for a significant acceleration of calculation times compared to conventional CPU solvers, not only saving time but also leading to considerable energy savings. With an increasing variety of models in the GPU solver, engineers are now able to perform complex calculations in record time, enabling not only precise results but also comprehensive parametric studies.
In addition to combustion and turbulence models, other physical aspects can, of course, be added: radiation and its effect on reaction rates, for example, play a central role. The same applies to coupled thermal simulations to determine the loads on mechanical components, or acoustic simulations. Furthermore, injection or spray atomization is an exciting CFD topic for which advanced modeling approaches, such as the VOF-to-DPM model, are available. To increase confidence in the calculation results, many comparisons with experiments are also referenced [3].
Do you want to deepen your understanding of all these phenomena? Participate in a special CADFEM training, for example on combustion [4] or turbulence [5], as modern simulation methods open up great opportunities for engineers to achieve faster, more accurate, and more efficient simulations, especially for the use of hydrogen in burners, turbines, or furnaces. Does your burner also operate with hydrogen instead of natural gas? Find out with simulation and CADFEM!
Written by CADFEM Germany GmbH