TL;DR
This paper reviews the development and validation of the Large Eddy Simulation (LES) coupled with the steady flamelet/progress variable (FPV) approach for simulating turbulent combustion. This method effectively captures complex phenomena like local extinction and re-ignition, proving valuable for understanding and predicting combustion in various scenarios, from fundamental flames to gas turbine combustors.
What Problem Does This Paper Address?
Simulating turbulent combustion is a significant challenge due to the intricate interplay between fluid dynamics and chemical reactions. While Large Eddy Simulation (LES) is adept at capturing large-scale turbulent structures, accurately modeling the sub-grid scale chemical processes, especially phenomena like flame extinction and re-ignition, remains difficult. Traditional combustion models often struggle with the complexity and stiffness of chemical kinetics. The paper addresses this by presenting and reviewing the LES-FPV approach, which aims to provide a more accurate and computationally efficient way to model these complex combustion phenomena.
Methods
The core of the paper revolves around the Large Eddy Simulation coupled with the steady flamelet/progress variable (LES-FPV) approach. Here's a breakdown:
- Large Eddy Simulation (LES): This is the underlying fluid dynamics solver. LES directly resolves the large, energy-containing turbulent eddies while modeling the smaller, sub-grid scale (SGS) turbulence. This is generally more accurate for complex flows than traditional Reynolds-Averaged Navier-Stokes (RANS) methods.
- Flamelet Model: This approach simplifies the complex chemistry by assuming that the reaction zones in turbulent flames are thin, locally one-dimensional structures called flamelets. These flamelets are assumed to be in a steady state, meaning their structure is primarily determined by the local mixing rate (represented by the mixture fraction and its dissipation rate) rather than the instantaneous flow dynamics.
- Progress Variable: To account for phenomena beyond simple mixing, a "progress variable" is introduced. This variable represents the overall advancement of the chemical reactions. By using the progress variable alongside the mixture fraction, the model can describe more complex combustion behaviors like extinction and re-ignition, which are not captured by purely mixing-based flamelet models.
- Steady Flamelet/Progress Variable (FPV) Approach: This combines the flamelet concept with the progress variable. The chemical species concentrations and temperature are expressed as functions of the mixture fraction and the progress variable. This significantly reduces the computational cost compared to solving full chemical kinetics.
- LES-FPV: The LES solver provides the filtered values of the mixture fraction and its dissipation rate, which are then used to look up the corresponding chemical species and temperature from pre-computed flamelet databases. The progress variable is also transported and modeled within the LES framework.
- Extensions: The paper also discusses extensions to the basic FPV model, such as incorporating sub-grid scale fluctuations of the progress variable (FPV-Cvar) and handling multi-component mixing (two mixture fractions) for more complex fuel mixtures and burner configurations.
Datasets
The paper extensively uses experimental data from various well-established turbulent flames and burner configurations for validation. These include:
- Sandia Series Flames (C, D, E, F): These are standard non-premixed jet flames used for validating combustion models, particularly for phenomena like extinction and re-ignition.
- Sydney Bluff-Body Flames: These flames, often used to mimic gas turbine combustor conditions, provide data for both non-reacting and reacting flows, including detailed measurements of velocity, temperature, and species concentrations.
- JVC Flames: These flames, characterized by vitiated co-flow, are designed to simulate the conditions of exhaust gas recirculation in combustion.
- JHC Flames: These three-stream jet flames allow for independent control of co-flow oxygen content and temperature, enabling detailed studies of low-oxygen combustion.
- PPJB Flames: Piloted Premixed Jet Burner flames, used to study premixed combustion in a simplified geometry.
- Dimethyl Ether (DME) Flames: These flames are used to investigate the combustion of more complex fuels.
The paper does not introduce new benchmark datasets but rather validates the LES-FPV approach against existing experimental data.
Key Insights and Ideas
- Bridging the Gap: The FPV approach effectively bridges the gap between the fluid dynamics captured by LES and the complex chemistry of combustion, particularly for phenomena like local extinction and re-ignition.
- Beyond Simple Mixing: Introducing the progress variable allows the model to go beyond simple mixing-controlled combustion and capture the dynamic evolution of chemical reactions.
- S-Curve Significance: The paper highlights the importance of the "S-curve" (representing the relationship between reaction rate and dissipation rate) in understanding ignition and extinction points. The FPV approach, by using the full S-curve, can better capture phenomena like local extinction and re-ignition compared to models that only consider one branch of the curve.
- Computational Efficiency: By pre-computing flamelet databases, the LES-FPV approach significantly reduces the computational cost associated with solving detailed chemical kinetics in real-time within an LES simulation.
- Versatility: The LES-FPV method has demonstrated its capability to simulate a wide range of turbulent flames, from simple jet flames to more complex configurations found in gas turbine combustors.
Key Results
The paper presents a comprehensive review of numerous validation studies. Key qualitative and quantitative findings include:
- Accurate Prediction of Flame Structures: The LES-FPV approach accurately predicts mean and instantaneous temperature fields, velocity profiles, and species mass fractions (e.g., H2O, CO2, OH, CO) for various flame types, including non-premixed, partially premixed, lifted, and swirl flames.
- Capturing Extinction and Re-ignition: The method successfully captures phenomena like local extinction and re-ignition in flames like the Sandia E flame, which are crucial for understanding flame stability.
- Gas Turbine Combustor Simulation: The approach has been applied to simulate gas turbine combustors, showing good agreement with experimental data for temperature and species distributions.
- Understanding Basic Combustion Phenomena: The LES-FPV method has been used to investigate fundamental combustion processes such as auto-ignition in lifted flames and the transition to MILD combustion.
- Improved Predictions: In many cases, the LES-FPV approach provides more accurate predictions than simpler models, such as steady flamelet models without progress variables. For example, it better predicts the location of OH peaks and the overall flame structure.
Limitations and Future Work
- Non-premixed Bias: The paper acknowledges that the LES-FPV approach is fundamentally derived from non-premixed flamelet models. While it shows good performance for some partially premixed flames, further improvements are needed for simulating typical partially premixed flames where mixing rates are not as high.
- Lewis Number Assumption: In some simulations, assuming a Lewis number of 1 for all species can lead to discrepancies, particularly in predicting reaction progress at downstream locations.
- Sub-grid Scale Modeling: While the FPV-Cvar extension accounts for sub-grid scale progress variable fluctuations, the modeling of sub-grid scale turbulence-chemistry interactions in general could still be an area for improvement.
- Complex Fuels: For complex fuels like Dimethyl Ether (DME), the model may not accurately capture all aspects of the pyrolysis process, leading to discrepancies in certain species concentrations (e.g., formaldehyde).
Future work will likely focus on extending the LES-FPV approach to handle more complex fuel mixtures, further refining sub-grid scale modeling for turbulence-chemistry interactions, and applying it to even more complex industrial combustion systems.
Summary updated