Dr. Steven (Yu) Xia, PhD CEng MIMechE

The demand for carbon-free fuels such as NH3 and H2 has been growing due to stricter environmental policies on carbon emissions. Since ammonia is easier to store and transport, it is considered an alternative fuel for marine, aviation, and gas turbine industries. However, the narrow flammability, low reactivity, and potential emissions of NOx are important challenges that need to be overcome if ammonia is to be used as a practical fuel for industrial use. Recent works have provided experimental… Show more

The demand for carbon-free fuels such as NH3 and H2 has been growing due to stricter environmental policies on carbon emissions. Since ammonia is easier to store and transport, it is considered an alternative fuel for marine, aviation, and gas turbine industries. However, the narrow flammability, low reactivity, and potential emissions of NOx are important challenges that need to be overcome if ammonia is to be used as a practical fuel for industrial use. Recent works have provided experimental data on the characteristics of NH3 flames under premixed and non-premixed combustion, focusing on flame speed, turbulence-chemistry interaction, and especially the dissociation/cracking of NH3 into H2 and N2. These measurements have encouraged the use of CFD for the simulation and scaling of practical ammonia systems, evaluating the performance of different numerical models for NH3 combustion. In this study, LES is utilized to predict a series of NH3/H2/N2 non-premixed flames stabilized on a bluff body burner. The accuracy of the FGM combustion model combined with LES has been examined, using a commercial CFD solver. The different cracking levels of NH3 into H2 and N2 lead to simulation cases ranging from an H2/N2 flame (i.e., “fully cracked”), to a flame with 50% NH3 (i.e., “half cracked”). The different flame shapes, temperature distributions, and NOx emissions have been simulated, generally matching well experimental data. The developed numerical settings and workflows may be applied to simulating more complex combustors which use NH3 as a fuel. Show less

Exergy based analysis for thermodynamic systems has become increasingly important in understanding the efficiency of energy-work conversion. Entropy generation, associated with the irreversibility in practical processes, decreases the available useful work, and the mechanisms responsible for the production differ significantly between premixed and nonpremixed combustion. In addition, unsteady combustion generates temperature fluctuations and when accelerated produces entropy waves – referred to… Show more

Exergy based analysis for thermodynamic systems has become increasingly important in understanding the efficiency of energy-work conversion. Entropy generation, associated with the irreversibility in practical processes, decreases the available useful work, and the mechanisms responsible for the production differ significantly between premixed and nonpremixed combustion. In addition, unsteady combustion generates temperature fluctuations and when accelerated produces entropy waves – referred to as indirect noise. Prior research by the authors have considered the dynamics of entropy generation purely due to heat release for both premixed and non-premixed flames. However, for non-premixed flames where the diffusion-controlled processes are important, transport terms must be considered for consistency. In this work, the entropy production mechanisms due to different sources are analyzed, firstly, by doing a mathematical study on the entropy transport equation and deriving the key forms of the source terms. Then reacting flow simulations are performed for the Sandia D flame, a well-documented configuration, for the numerical analysis of the transport equation. A range of operating conditions have been considered to illustrate the relative importance of different entropy source terms. Overall entropy production rate budget is calculated, and trends are observed for different operating conditions. In addition, across different sections of the flame, local entropy production is analyzed. Subsequently, reacting flow simulations are performed for the classical Burke-Schumann flame, a laminar nonpremixed flame, and the same analysis is carried out to completely gauge the trends of entropy production in non-premixed flames. Show less

The optimization of Venturi mixers in burners is critical for enhancing combustion efficiency and minimizing emissions. In this study, we utilize the adjoint method to analyze and refine the design of a Venturi mixer. Our numerical simulations integrate the species transport equation with the Eddy Dissipation Model (EDM) for reacting flow and the generalized k-omega (GEKO) model to simulate turbulence. By solving adjoint equations, we effectively compute the shape sensitivity for various… Show more

The optimization of Venturi mixers in burners is critical for enhancing combustion efficiency and minimizing emissions. In this study, we utilize the adjoint method to analyze and refine the design of a Venturi mixer. Our numerical simulations integrate the species transport equation with the Eddy Dissipation Model (EDM) for reacting flow and the generalized k-omega (GEKO) model to simulate turbulence. By solving adjoint equations, we effectively compute the shape sensitivity for various observables, including pressure drop, outlet fuel variance/uniformity deviation index, air and fuel mass flow rates, and outlet CO mass fraction. The shape sensitivity analysis uncovers the interplay between the observables and the appropriate weights for multiple objective optimizations. Subsequently, we perform gradient-based optimizations to enhance the mixer's performance, employing both shape sensitivity and mesh morphing techniques. We conduct a series of case studies focusing on both cold and reacting flows. The optimization of cold flow provides an in-depth exploration of various optimization strategies, encompassing single-objective and multi-objective optimization with diverse weight combinations. Following this, the optimization of reacting flow enhances the mixer's functionality under combustion conditions, emphasizing the reduction of emissions and the increase of combustion efficiency. Our findings showcase the potential of an adjoint-based optimization framework in designing Venturi mixers that are efficient and emit lower levels of pollutants. Show less

Hypersonic flows pose significant challenges in aerospace engineering and atmospheric sciences. While accurate physical modeling and numerical prediction of hypersonic flows are crucial for the design and analysis of this class of vehicles, the extreme conditions make this class of numerical simulations very difficult to converge to a stable and steady solution. The authors' recent study utilized a "two-temperature model" to simulate the hypersonic flow and thermal features of the FIRE-II… Show more

Hypersonic flows pose significant challenges in aerospace engineering and atmospheric sciences. While accurate physical modeling and numerical prediction of hypersonic flows are crucial for the design and analysis of this class of vehicles, the extreme conditions make this class of numerical simulations very difficult to converge to a stable and steady solution. The authors' recent study utilized a "two-temperature model" to simulate the hypersonic flow and thermal features of the FIRE-II re-entry vehicle. This previous work helped in understanding important trends for surface heat flux, temperature field, and ionization of air, etc., at five operating points along the vehicle's re-entry trajectory. In the present work, further investigations have been performed on this vehicle by (1) comparing two different chemical reaction mechanisms; (2) applying full and partial catalytic wall (PCW) conditions for surface heat flux predictions. Both chemical mechanisms, the Park and the Gupta models, account for the recombination of the ionized species and have proven to better predict the heating trends on the surface of the capsule. The use of catalytic wall model also improves the prediction of surface heat flux on the vehicle across all the operating points compared to previous studies. While trends in heat flux are captured well for different altitudes, the absolute values and the match with the experiment still present some discrepancies. These discrepancies are in large part attributable to the presence of a large uncertainty in the exact, instantaneous flight conditions, in the assumption of steady-state conditions at each altitude as well as unknown catalytic conditions of the capsule's thermal protection system (TPS). The developed workflow and best practice settings from this study can be applied to more complex hypersonic applications. Show less

The optimization of Venturi mixers in burners is critical for enhancing combustion efficiency and minimizing emissions. In this study, we utilize the adjoint method to analyze and refine the design of a Venturi mixer. Our numerical simulations integrate the species transport equation with the Eddy Dissipation Model (EDM) for reacting flow and the generalized k-ω (GEKO) model to simulate turbulence. By solving adjoint equations, we effectively compute the shape sensitivity for various… Show more

The optimization of Venturi mixers in burners is critical for enhancing combustion efficiency and minimizing emissions. In this study, we utilize the adjoint method to analyze and refine the design of a Venturi mixer. Our numerical simulations integrate the species transport equation with the Eddy Dissipation Model (EDM) for reacting flow and the generalized k-ω (GEKO) model to simulate turbulence. By solving adjoint equations, we effectively compute the shape sensitivity for various observables, including pressure drop, outlet fuel variance/uniformity deviation index, air and fuel mass flow rates, and outlet CO mass fraction. The shape sensitivity analysis uncovers the interplay between the observables and the appropriate weights for multiple objective optimizations. Subsequently, we perform gradient-based optimizations to enhance the mixer's performance, employing both shape sensitivity and mesh morphing techniques. We conduct a series of case studies focusing on both cold and reacting flows. The optimization of cold flow provides an in-depth exploration of various optimization strategies, encompassing single-objective and multi-objective optimization with diverse weight combinations. Following this, the optimization under reacting flow conditions improves the fuel/air mixing, leading to the increase of combustion efficiency and hence the reduction of CO emissions. Our findings showcase the potential of an adjoint-based optimization framework in designing Venturi mixers that are efficient and emit lower levels of pollutants. Show less

The demand for carbon-free fuels such as ammonia (NH3) and hydrogen (H2) has been growing rapidly due to stricter environmental policies on carbon emissions. Since ammonia is easier to store and transport, it is considered the leading alternative fuel for marine, aviation, and gas turbine industries. However, the narrow flammability, low reactivity, and potential emissions of nitrogen oxides (NOx) are important challenges that need to be overcome if ammonia is to be used as a practical fuel for… Show more

The demand for carbon-free fuels such as ammonia (NH3) and hydrogen (H2) has been growing rapidly due to stricter environmental policies on carbon emissions. Since ammonia is easier to store and transport, it is considered the leading alternative fuel for marine, aviation, and gas turbine industries. However, the narrow flammability, low reactivity, and potential emissions of nitrogen oxides (NOx) are important challenges that need to be overcome if ammonia is to be used as a practical fuel for industrial use. Recent works have provided important experimental data on the characteristics of NH3 flames under premixed and non-premixed combustion modes, focusing on flame speed, turbulence-chemistry interaction, and especially the dissociation/ cracking of NH3 into H2 and N2. These measurements have encouraged the use of Computational Fluid Dynamics (CFD) for the simulation and scaling of practical ammonia systems, evaluating the performance of different numerical models for NH3 combustion. In this study, Large Eddy Simulation (LES) is utilized to predict a series of NH3/H2/N2 non-premixed flames stabilized on a bluff body burner. The accuracy of the Flamelet Generated Manifold (FGM) combustion model combined with LES has been examined, using a commercial CFD solver. The different cracking levels of NH3 into H2 and N2 lead to simulation cases ranging from an H2/N2 flame (i.e., “fully cracked”), to a flame with 50% NH3 (i.e., “half cracked”). The different flame shapes, temperature distributions, NOx and unburnt NH3 emissions have all been simulated, generally matching well the available experimental data. The developed model settings and numerical workflows may be applied to simulating more complex combustors which use NH3 as a fuel. Show less

The demand for carbon-free fuels such as ammonia (NH3) and hydrogen (H2) has been growing rapidly in the last decade due to stricter environmental policies being applied on carbon emission. The storage and transport of NH3 in the liquid form, at a relatively low tank pressure, are comparatively easier than those of H2. As a result, NH3 is considered an alternative H2-carrier fuel and has become a potential energy provider for marine, aviation, and gas turbine industries. However, the narrow… Show more

The demand for carbon-free fuels such as ammonia (NH3) and hydrogen (H2) has been growing rapidly in the last decade due to stricter environmental policies being applied on carbon emission. The storage and transport of NH3 in the liquid form, at a relatively low tank pressure, are comparatively easier than those of H2. As a result, NH3 is considered an alternative H2-carrier fuel and has become a potential energy provider for marine, aviation, and gas turbine industries. However, the narrow flammability, low reactivity, and potential emission of nitrogen oxides (NOx), have all been important challenges for ammonia to be used as a practical fuel in industrial combustors. The goal of this work is to perform LES-FGM simulations for different ammonia cracking cases (flames). The present LES/FGM simulations predict a series of NH3/H2/N2 blended flames quite well. These include mean temperature, NOx and unburnt NH3 mass & volume fractions.The increase of NH3 and decrease of jet velocity shorten the flame and reduce the flow temperature, bringing down the mixture reactivity due to reduction of H2. A radiation model is being included to get even better thermal behavior of the flames. As a next step, solving transport equations for slow-evolving species (e.g., Nox) within FGM is being implemented. Sensitivity analyses are underway to provide full diagnostics and design capabilities for this fuel of the future. The developed workflow will be useful for NH3-fueled burner design & certification. Show less

In this paper, the numerical model of Eulerian Wall Film (EWF) implemented within Ansys Fluent® has been validated based on two canonical cases. The first one refers to the experiments performed by Hu et al. [1], where the heat transfer during the water film evaporation on a vertical plate has been studied. The average heat flux and water film evaporative ratio match the available experimental data well. The second case relates to the experiments done by Ambrosini et al. [2], also known as the… Show more

In this paper, the numerical model of Eulerian Wall Film (EWF) implemented within Ansys Fluent® has been validated based on two canonical cases. The first one refers to the experiments performed by Hu et al. [1], where the heat transfer during the water film evaporation on a vertical plate has been studied. The average heat flux and water film evaporative ratio match the available experimental data well. The second case relates to the experiments done by Ambrosini et al. [2], also known as the "CONAN" case, where the wall film condensation in the presence of non-condensable substances has been studied under different steam mass fraction and velocity conditions. The surface heat flux and condensate mass flow rate also match the experimental measurements. The present numerical workflow using EWF could be applied to more complex devices for liquid film evaporation and condensation simulations. Show less

This study investigates the transport of entropy waves in a industrial complex geometry using Large Eddy Simulation and linear analysis. The LES include a transport equation for a passive scalar, which models entropy. The fluctuations in the passive scalar (or entropy waves) are introduced at the temporally averaged flame location via a harmonically fluctuating source term. The advection and diffusion of the fluctuations in passive scalar are investigated by time stepping the LES. Subsequently,… Show more

This study investigates the transport of entropy waves in a industrial complex geometry using Large Eddy Simulation and linear analysis. The LES include a transport equation for a passive scalar, which models entropy. The fluctuations in the passive scalar (or entropy waves) are introduced at the temporally averaged flame location via a harmonically fluctuating source term. The advection and diffusion of the fluctuations in passive scalar are investigated by time stepping the LES. Subsequently, the linear analysis of the entropy wave convection is conducted. The transport equation of the passive scalar is linearized around the temporal mean state, which is obtained by the LES, and transferred into frequency domain, which leads to a transport equation of passive scalar fluctuations (entropy waves). The diffusion of entropy waves is modeled using a turbulent diffusivity, which in this study is determined by a k-epsilon model. The linearized transport equation of the passive scalar is solved for the same frequencies, which are investigated in the LES, using the finite element approach. The results obtained by the linearized approach is in good agreement with the LES, which demonstrates that it is well suited to address the transport of entropy waves in real world combustors. Show less

Hypersonic flows pose significant challenges in aerospace engineering and atmospheric sciences. Accurate modeling and prediction of hypersonic flows are crucial for various applications, including space exploration, re-entry atmospheric vehicles, and high-speed propulsion systems. Kumar et al. [1] in their previous work attempted to validate the two-temperature modeling approach against the measured heat flux data of FIRE-II re-entry vehicle for five different trajectory points. This previous… Show more

Hypersonic flows pose significant challenges in aerospace engineering and atmospheric sciences. Accurate modeling and prediction of hypersonic flows are crucial for various applications, including space exploration, re-entry atmospheric vehicles, and high-speed propulsion systems. Kumar et al. [1] in their previous work attempted to validate the two-temperature modeling approach against the measured heat flux data of FIRE-II re-entry vehicle for five different trajectory points. This previous work helped in establishing a few important trends regarding heat flux, the size of the recirculation region, and the extent of ionization with respect to the Reynolds number In the current work, two different reaction mechanisms, Gupta model [2] and Park Model [3] are further evaluated for their applicability at five different trajectory points along the re-entry trajectory . Additionally, the heat flux predictions on the afterbody have been improved by using partial catalytic wall models as proposed in the references [7-9]. The two chemical reacting models allow the recombination of ionized atoms to form molecules and better predict the heat exchange due to reactions. An improvement in the heat flux prediction on the afterbody is consistently seen across all the five trajectory points of the FIRE-II measurements. Show less

As hydrogen is gaining its popularity as an alternative source of energy, there is a strong focus on understanding the hazards related to transport of hydrogen. Nowadays, the most popular means to transport hydrogen is to store it in highly pressurized cylinders. The current work focuses on numerically capturing the auto-ignition pattern during a sudden release of hydrogen into the atmosphere for different pressurized conditions. This paper attempts to understand the auto-ignition of hydrogen… Show more

As hydrogen is gaining its popularity as an alternative source of energy, there is a strong focus on understanding the hazards related to transport of hydrogen. Nowadays, the most popular means to transport hydrogen is to store it in highly pressurized cylinders. The current work focuses on numerically capturing the auto-ignition pattern during a sudden release of hydrogen into the atmosphere for different pressurized conditions. This paper attempts to understand the auto-ignition of hydrogen using the latest computational fluid Dynamics (CFD) techniques, simulating the 3D turbulent structures, the hydrogen diffusion into the air, and performing the analysis of different reaction mechanisms on predicting whether the auto-ignition would happen or not, and if it happens whether it will be sustained or later quenched. Numerical simulations are carried out at three different cylinder pressures (50 bar, 100 bar and 250 bar), using selected reaction mechanisms with their corresponding transport and thermochemical properties that can capture the auto-ignition, such as the Li Mechanism, the Connaire Mechanism, and the Ansys Model Fuel Library (MFL), etc. To correctly capture the auto-ignition phenomenon, it is vital to accurately predict the diffusion of pressurized hydrogen into the ambient, which may be achieved by using an appropriate turbulence model such as Large Eddy Simulation (LES). Secondly, the mesh grid used needs to be sufficiently fine to capture the small turbulent eddies which enhance the underlying hydrogen flow fluctuations. Finally, the material properties of the hydrogen as a real gas must also be carefully defined. Show less

This work utilizes the Ansys Fluent® solver for modeling reacting flow and two-way FSI, applied to a laboratory-scale 3D methane/air burner. The burner features a bluff-body stabilized, lean partially premixed flame experiencing strong thermo-acoustic oscillations. In experiments, a thin steel liner is installed around the main combustion chamber, which heavily interacts with the flame and flow field, to produce large amplitude structural deformation. An unsteady RANS approach uses the Shear… Show more

This work utilizes the Ansys Fluent® solver for modeling reacting flow and two-way FSI, applied to a laboratory-scale 3D methane/air burner. The burner features a bluff-body stabilized, lean partially premixed flame experiencing strong thermo-acoustic oscillations. In experiments, a thin steel liner is installed around the main combustion chamber, which heavily interacts with the flame and flow field, to produce large amplitude structural deformation. An unsteady RANS approach uses the Shear Stress Transport (SST) turbulence model and a Flamelet Generated Manifold (FGM) combustion model to predict the turbulent reacting flow within the burner. The solver has a built-in Finite Element (FE) structure model, which simultaneously solves the displacement equations and the turbulent reacting flow governing equations. This way, a fully coupled 2-way FSI simulation is carried out to predict thermo-acoustic instabilities in the burner and associated wall deformations. Overall pressure oscillation and liner wall displacement (frequency and amplitude) are in good agreement with experimental data with results for different operating conditions. Although the studied combustor is simplistic, the established 2-way FSI workflow will support commercial gas turbine combustor design and prognosis. Show less

Popular turbulent combustion models like flamelet generated manifold (FGM) have been extensively used for hydrocarbon flames. However, their applicability, usability, and accuracy need to be evaluated in modeling H2 flames. The development of best practices for H2 combustion modeling using FGM requires an understanding of how the progress variable is to be defined, the mesh requirements, any numerical challenges due to faster burning rate, and the sensitivity of FGM to change in input… Show more

Popular turbulent combustion models like flamelet generated manifold (FGM) have been extensively used for hydrocarbon flames. However, their applicability, usability, and accuracy need to be evaluated in modeling H2 flames. The development of best practices for H2 combustion modeling using FGM requires an understanding of how the progress variable is to be defined, the mesh requirements, any numerical challenges due to faster burning rate, and the sensitivity of FGM to change in input parameters. This is the motivation of the current work in which a turbulent lifted H2/N2 jet in vitiating co-flow environment is investigated using the FGM model with large eddy simulation (LES). Due to hot vitiated co-flow, the primary mechanism of stabilization is auto-ignition followed by a premixed flame. The FGM manifold is created from steady-state flamelet solution along with one quenching flamelet and is typically associated with one underlying flamelet configuration, premixed or diffusion. In addition to the H2 as a fuel, this flame poses two other modeling challenges. First is an auto-ignition event, a transient chemistry-driven phenomenon, and second is the existence of multiple regimes, diffusion at auto-ignition location, and premixed in the post flame. In this work, four simulations are done by varying the co-flow temperature from 1000 K to 1045 K. The progress variable is computed as the sum of H2O and HO2 with different weights. The FGM model can predict the characteristics of the flame by showing a lifted flame. It also accurately predicted the trend in the flame lift-off distance with a change in the co-flow temperature. The current results are also compared for mixture fraction, temperature, and OH mass fraction at multiple locations, which have also been correctly captured. Show less

A turbulent lifted H2/N2 jet flame in a vitiating co-flow environment is numerically investigated, using the Flamelet Generated Manifold (FGM) combustion model with Large Eddy Simulations (LES). Due to the hot vitiated H2/air co-flow, the primary stabilization mechanism is the autoignition followed by a premixed flame. In addition to the H2 as a fuel, this flame poses two other modeling challenges: (i) the autoignition, which is a transient chemistry-driven phenomenon; (ii) the existence of… Show more

A turbulent lifted H2/N2 jet flame in a vitiating co-flow environment is numerically investigated, using the Flamelet Generated Manifold (FGM) combustion model with Large Eddy Simulations (LES). Due to the hot vitiated H2/air co-flow, the primary stabilization mechanism is the autoignition followed by a premixed flame. In addition to the H2 as a fuel, this flame poses two other modeling challenges: (i) the autoignition, which is a transient chemistry-driven phenomenon; (ii) the existence of multiple combustion regimes, e.g., diffusion at autoignition location but premixed in the post flame. A series of LES/FGM simulations are completed in this work by reducing the co-flow temperature from 1045 K to 1000 K. The FGM model can predict the characteristics of the flame by showing a lifted flame. It also accurately predicts the trend in the flame lift-off distance with a change in the co-flow temperature. The current results are compared for mixture fraction, temperature, and OH mass fraction at multiple locations, which have also been correctly captured. It is noted that for a high co-flow temperature (and hence a low lift-off distance), the flame's lift-off is highly sensitive to the inlet boundary conditions and the mesh resolution near the jet entry. A relatively coarse mesh is used for all the simulations, which is generated using a careful strategy that not only resolves the jet instabilities near the fuel inlet, but also keeps the overall mesh count low and allows for a large computational time step. A systematic sensitivity analysis on the computational speed is also performed. This work provides some useful guidelines in simulating the H2 diluted flames using the FGM model, which may be valuable to the gas turbine industry. Show less

This work utilizes the Ansys Fluent® solver for modeling reacting flow and two-way FSI, applied to a laboratory-scale 3D methane/air burner. The burner features a bluff body stabilized, lean partially premixed flame experiencing strong thermo-acoustic oscillations. In experiments, a thin steel liner is installed around the main combustion chamber, which heavily interacts with the flame and flow field, to produce large amplitude structural deformation. An unsteady RANS approach uses the Shear… Show more

This work utilizes the Ansys Fluent® solver for modeling reacting flow and two-way FSI, applied to a laboratory-scale 3D methane/air burner. The burner features a bluff body stabilized, lean partially premixed flame experiencing strong thermo-acoustic oscillations. In experiments, a thin steel liner is installed around the main combustion chamber, which heavily interacts with the flame and flow field, to produce large amplitude structural deformation. An unsteady RANS approach uses the Shear Stress Transport (SST) turbulence model and a Flamelet Generated Manifold (FGM) combustion model to predict the turbulent reacting flow within the burner. The solver has a built-in Finite Element (FE) structure model, which simultaneously solves the displacement equations and the turbulent reacting flow governing equations. This way, a fully coupled 2-way FSI simulation is carried out to predict thermo-acoustic instabilities in the burner and associated wall deformations. Overall pressure oscillation and liner wall displacement (frequency and amplitude) are in good agreement with experimental data with results for different operating conditions. Although the studied combustor is simplistic, the established 2-way FSI workflow will support commercial gas turbine combustor design and prognosis. Show less

In this paper, a dynamic adaptive mesh refinement method is used in conjunction with a hybrid scale-resolving turbulence model to solve industrial combustion problems. The objective of the adaption method is to track and resolve characteristic turbulent structures arising from swirlers, pilot injectors and flame propagation in industrial burner configurations. By employing Polyhedral Unstructured Mesh Adaption (PUMA) within Ansys Fluent solver, local regions of mesh are refined to capture… Show more

In this paper, a dynamic adaptive mesh refinement method is used in conjunction with a hybrid scale-resolving turbulence model to solve industrial combustion problems. The objective of the adaption method is to track and resolve characteristic turbulent structures arising from swirlers, pilot injectors and flame propagation in industrial burner configurations. By employing Polyhedral Unstructured Mesh Adaption (PUMA) within Ansys Fluent solver, local regions of mesh are refined to capture gradients in temperature, velocity and other key variables. For Scale-Resolving Simulations (SRS), highly refined meshes are required to resolve a sufficient range of turbulent scales. In this work, a strategy is proposed to evaluate the scale-resolving quality of the mesh and to refine it dynamically in a transient simulation. The condition used for adapting the mesh is based on the gradients of key variables such as temperature and velocity, whilst the large-scale eddies are resolved using an approach based on the LES mesh resolution index. This strategy is then applied to a series of test cases (a piloted dilution flame, a bluffbody premixed flame and a swirl stabilized flame), using the hybrid Stress-Blended Eddy Simulation (SBES) turbulence model and a Flamelet Generated Manifold (FGM) combustion model. The numerical results are compared with available experimental data, and the accuracy of the solutions is discussed. Show less

Accurate numerical prediction of surface heat transfer in the presence of film cooling within aero-engine sub-components, such as blade effusion holes and combustor liners, has long been a goal of the aero-engine industry. It requires accurate simulation of the turbulent mixing and reaction processes between freestream and the cooling flow. In this study, the Stress Blended Eddy Simulation (SBES) turbulence model is used together with the Flamelet Generated Manifold (FGM) combustion model to… Show more

Accurate numerical prediction of surface heat transfer in the presence of film cooling within aero-engine sub-components, such as blade effusion holes and combustor liners, has long been a goal of the aero-engine industry. It requires accurate simulation of the turbulent mixing and reaction processes between freestream and the cooling flow. In this study, the Stress Blended Eddy Simulation (SBES) turbulence model is used together with the Flamelet Generated Manifold (FGM) combustion model to calculate the surface heat flux upstream and downstream of an effusion cooling hole. The SBES model employs a blending function to automatically switch between RANS and LES based on the local flow features, and thus significantly reduces the computational cost compared to a full LES simulation. All simulations are run using ANSYS Fluent, a commercial finite-volume CFD solver. The test case corresponds to an experimental rig run at MIT, which is essentially a flat plate brushed by a uniform freestream of argon with ethylene seeded inside, and is cooled by either a reacting air or non-reacting nitrogen jet inclined at 35 degrees to the freestream. Calculations are performed for both reacting and non-reacting jet cooling cases across a range of jet-to-stream blowing ratios, and compared with the experimental data. The effects of mesh resolution are also investigated. Calculations are also performed across a range of Damkohler number (i.e. flow to chemical time ratio) from zero to 30, with unity blowing ratio, and the differences in the maximum surface heat flux magnitude in the reacting and non-reacting cases at a specific location downstream of the hole are investigated. Results from these analyses show good correlation with the experimental heat flux data upstream and downstream of the cooling hole, including the heat flux augmentation due to local reaction. Results from the Damkohler number sweep also show a good match with the experimental data across the range investigated. Show less

Accurate numerical prediction of surface heat transfer in the presence of film cooling within aero-engine sub-components, such as blade effusion holes and combustor liners, has long been a goal of the aero-engine industry. It requires accurate simulation of the turbulent mixing and reaction processes between freestream and the cooling flow. In this study, the Stress Blended Eddy Simulation (SBES) turbulence model is used together with the Flamelet Generated Manifold (FGM) combustion model to… Show more

Accurate numerical prediction of surface heat transfer in the presence of film cooling within aero-engine sub-components, such as blade effusion holes and combustor liners, has long been a goal of the aero-engine industry. It requires accurate simulation of the turbulent mixing and reaction processes between freestream and the cooling flow. In this study, the Stress Blended Eddy Simulation (SBES) turbulence model is used together with the Flamelet Generated Manifold (FGM) combustion model to calculate the surface heat flux upstream and downstream of an effusion cooling hole. The SBES model employs a blending function to automatically switch between RANS and LES based on the local flow features, and thus significantly reduces the computational cost compared to a full LES simulation. All simulations are run using ANSYS Fluent, a commercial finite-volume CFD solver. The test case corresponds to an experimental rig run at MIT, which is essentially a flat plate brushed by a uniform freestream of argon with ethylene seeded inside, and is cooled by either a reacting air or non-reacting nitrogen jet inclined at 35 degrees to the freestream. Calculations are performed for both reacting and non-reacting jet cooling cases across a range of jet-to-stream blowing ratios, and compared with the experimental data. The effects of mesh resolution are also investigated. Calculations are also performed across a range of Damkohler number (i.e. flow to chemical time ratio) from zero to 30, with unity blowing ratio, and the differences in the maximum surface heat flux magnitude in the reacting and non-reacting cases at a specific location downstream of the hole are investigated. Results from these analyses show good correlation with the experimental heat flux data upstream and downstream of the cooling hole, including the heat flux augmentation due to local reaction. Results from the Damkohler number sweep also show a good match with the experimental data across the range investigated. Show less

This paper describes numerical predictions of the forced heat release responses of a lean premixed bluff-body stabilized ethylene/air flame, using the hybrid Stress-Blended Eddy Simulation (SBES) turbulence model and the Flamelet Generated Manifold (FGM) combustion model. The unforced flow field and flame structures are firstly simulated and validated against available experimental data. The forced heat release rate responses of the flame to upstream harmonic velocity fluctuations are then… Show more

This paper describes numerical predictions of the forced heat release responses of a lean premixed bluff-body stabilized ethylene/air flame, using the hybrid Stress-Blended Eddy Simulation (SBES) turbulence model and the Flamelet Generated Manifold (FGM) combustion model. The unforced flow field and flame structures are firstly simulated and validated against available experimental data. The forced heat release rate responses of the flame to upstream harmonic velocity fluctuations are then computed and accounted for using the weakly nonlinear Flame Describing Function (FDF), which again match well the experimental measurements. These findings indicate that the use of SBES and FGM models can accurately predict the forced flame responses and the resultant FDFs, which can be applied to predicting thermoacoustic instabilities in real gas turbine combustors. Show less

This thesis numerically predicts thermoacoustic instability in two typical gas turbine combustors. Incompressible large eddy simulation (LES) is used as the main numerical tool to simulate the heat release response of a flame to upstream acoustic perturbations, resulting in linear and nonlinear flame response models. These flame models are then combined with analytical acoustic wave models to predict thermoacoustic instability, and for those unstable operations, the resulting nonlinear limit… Show more

This thesis numerically predicts thermoacoustic instability in two typical gas turbine combustors. Incompressible large eddy simulation (LES) is used as the main numerical tool to simulate the heat release response of a flame to upstream acoustic perturbations, resulting in linear and nonlinear flame response models. These flame models are then combined with analytical acoustic wave models to predict thermoacoustic instability, and for those unstable operations, the resulting nonlinear limit cycle oscillations. Other features relevant to thermoacoustic instability are also studied, including entropy wave transport within the combustor flow fields, and the axial variation of the heat release response along a long flame length, etc. Show less

The forced flame responses in a pressurised gas turbine combustor are predicted using numerical reacting flow simulations. Two incompressible LES solvers are used, applying two combustion models and two reaction schemes (4-step and 15-step) at two operating pressures (3 bar and 6 bar). Although the combustor flow field is little affected by these factors, the flame length and heat release rate are found to depend on combustion model, reaction scheme and combustor pressure. The flame responses… Show more

The forced flame responses in a pressurised gas turbine combustor are predicted using numerical reacting flow simulations. Two incompressible LES solvers are used, applying two combustion models and two reaction schemes (4-step and 15-step) at two operating pressures (3 bar and 6 bar). Although the combustor flow field is little affected by these factors, the flame length and heat release rate are found to depend on combustion model, reaction scheme and combustor pressure. The flame responses to an upstream velocity perturbation are used to construct the flame describing functions (FDFs). The FDFs exhibit smaller dependence on the combustion model and reaction chemistry than the flame shape and mean heat release rate. The FDFs are validated by predicting combustor thermoacoustic stability at 3 bar and 6 bar and, for the unstable 6 bar case, also by predicting the frequency and oscillation amplitude of the resulting limit cycle oscillation. All of these numerical predictions are in very good agreement with experimental measurements. Show less

This work numerically investigates the impact of flame-to-flame interactions on the flame response to upstream acoustic perturbations in a turbulent swirling annular combustor. A section of three identical flames taken from the original annular rig is studied, using incompressible large eddy simulations (LES). The un-forced flow-field and flame behaviour are firstly simulated and validated by experimental data. A fixed harmonic velocity perturbation is then imposed upstream to each flame and… Show more

This work numerically investigates the impact of flame-to-flame interactions on the flame response to upstream acoustic perturbations in a turbulent swirling annular combustor. A section of three identical flames taken from the original annular rig is studied, using incompressible large eddy simulations (LES). The un-forced flow-field and flame behaviour are firstly simulated and validated by experimental data. A fixed harmonic velocity perturbation is then imposed upstream to each flame and the heat release rate response from the central flame is computed over time, which leads to the construction of the weakly nonlinear flame describing function (FDF). The effect of the separation distance between neighbouring flames (denoted S) on the magnitude and trend of the FDF is analysed, showing that a drop in S leads to an increase in FDF-gain at low but a decrease at high frequencies. The increase of perturbation amplitude results in a monotonic decrease of the FDF-gain regardless of the value of S. Show less

This article presents numerical prediction of a thermoacoustic limit cycle in an industrial gas turbine combustor. The case corresponds to an experimental high pressure test rig equipped with the full-scale Siemens SGT-100 combustor operated at two mean pressure levels of 3 bar and 6 bar. The Flame Transfer Function (FTF) characterising the global unsteady response of the flame to velocity perturbations is obtained for both operating pressures by means of incompressible Large Eddy Simulations… Show more

This article presents numerical prediction of a thermoacoustic limit cycle in an industrial gas turbine combustor. The case corresponds to an experimental high pressure test rig equipped with the full-scale Siemens SGT-100 combustor operated at two mean pressure levels of 3 bar and 6 bar. The Flame Transfer Function (FTF) characterising the global unsteady response of the flame to velocity perturbations is obtained for both operating pressures by means of incompressible Large Eddy Simulations (LES). A linear stability analysis is then performed by coupling the FTFs with a wave-based low order thermoacoustic network solver. All the thermoacoustic modes predicted at 3 bar pressure are stable; whereas one of the modes at 6 bar is found to be unstable at a frequency of 231 Hz, which agrees with the experiments. A weakly nonlinear stability analysis is carried out by combining the Flame Describing Function (FDF) predicted by LES with the low order thermoacoustic network solver. The frequency, mode shape and velocity amplitude corresponding to the predicted limit cycle at 209 Hz are used to compute the absolute pressure fluctuation amplitude in the combustor. The numerically reconstructed amplitude is found to be reasonably close to the measured dynamics. Show less

In the context of combustion noise and combustion instabilities, the transport of entropy perturbations through highly simplified turbulent flows has received much recent attention. This work performs the first systematic study into the transport of entropy perturbations through a realistic gas turbine combustor flow-field, exhibiting large-scale hydrodynamic flow features in the form of swirl, separation, recirculation zones and vortex cores, these being ubiquitous in real combustor flows. The… Show more

In the context of combustion noise and combustion instabilities, the transport of entropy perturbations through highly simplified turbulent flows has received much recent attention. This work performs the first systematic study into the transport of entropy perturbations through a realistic gas turbine combustor flow-field, exhibiting large-scale hydrodynamic flow features in the form of swirl, separation, recirculation zones and vortex cores, these being ubiquitous in real combustor flows. The reacting flow-field is simulated using low Mach number large eddy simulations, with simulations validated by comparison to available experimental data. A generic artificial entropy source, impulsive in time and spatially localized at the flame-front location, is injected. The conservation equation describing entropy transport is simulated, superimposed on the underlying flow-field simulation. It is found that the transport of entropy perturbations is dominated by advection, with both thermal diffusion and viscous production being negligible. It is furthermore found that both the mean flow-field and the large-scale unsteady flow features contribute significantly to advective dispersion — neither can be neglected. The time-variation of entropy perturbation amplitude at combustor exit is well-modelled by a Gaussian profile, whose dispersion exceeds that corresponding to a fully-developed pipe mean flow profile roughly by a factor of three. Finally, despite the attenuation in entropy perturbation amplitude caused by advective dispersion, sufficient entropy perturbation strength is likely to remain at combustor exit for entropy noise to make a meaningful contribution at low frequencies. Show less

This work numerically investigates the unsteady heat release rate response of a full-scale indus- trial gas turbine combustor to acoustic perturbations. The combustor contains a lean technically- premixed methane/air flame. Two large eddy simulation solvers are compared, the first being the in-house code BOFFIN which employs a reduced 15-step chemical reaction mechanism; the second is based on the open-source CFD toolbox OpenFOAM and applies both 2-step and 4-step reaction mechanisms. Both are… Show more

This work numerically investigates the unsteady heat release rate response of a full-scale indus- trial gas turbine combustor to acoustic perturbations. The combustor contains a lean technically- premixed methane/air flame. Two large eddy simulation solvers are compared, the first being the in-house code BOFFIN which employs a reduced 15-step chemical reaction mechanism; the second is based on the open-source CFD toolbox OpenFOAM and applies both 2-step and 4-step reaction mechanisms. Both are incompressible codes, exploiting the fact that the flame responds to hydrodynamic perturbations excited by the acoustics. For the unforced flow-field, the simula- tion results agree well with the experiments. The flame heat release rate responses are calculated by applying a harmonic forcing velocity upstream to the flame across two forcing amplitudes and eight forcing frequencies. The obtained frequency responses of flame are known as flame de- scribing functions, which are different between the solvers used, especially on their gains. This indicates that for combustors with industrial complexity, more detailed chemistry mechanisms may be necessary. The phase lags of the flame describing functions generally decrease linearly with forcing frequency, being almost independent of the forcing amplitude and the solver used. Show less

A coupled numerical approach is investigated for predicting combustion instability limit cycle characteristics when the combustor contains a long flame.

The thermoacoustic modes of a full scale industrial gas turbine combustor have been predicted numerically. The predictive approach combines low order network modelling of the acoustic waves in a simplified geometry, with a weakly nonlinear flame describing function, obtained from incompressible large eddy simulations of the flame region under upstream forced velocity perturbations, incorporating reduced chemistry mechanisms. Two incompressible solvers, each employing different numbers of… Show more

The thermoacoustic modes of a full scale industrial gas turbine combustor have been predicted numerically. The predictive approach combines low order network modelling of the acoustic waves in a simplified geometry, with a weakly nonlinear flame describing function, obtained from incompressible large eddy simulations of the flame region under upstream forced velocity perturbations, incorporating reduced chemistry mechanisms. Two incompressible solvers, each employing different numbers of reduced chemistry mechanism steps, are used to simulate the turbulent reacting flowfield to predict the flame describing functions. The predictions differ slightly between reduced chemistry approximations, indicating the need for more involved chemistry. These are then incorporated into a low order thermoacoustic solver to predict thermoacoustic modes. For the combustor operating at two different pressures, most thermoacoustic modes are predicted to be stable, in agreement with the experiments. The predicted modal frequencies are in good agreement with the measurements, although some mismatches in the predicted modal growth rates and hence modal stabilities are observed. Overall, these findings lend confidence in this coupled approach for real industrial gas turbine combustors. Show less

In gas turbines, unsteady combustion generates temperature fluctuations, which are often termed entropy waves. When these are accelerated through the turbine stages, they represent a significant source of noise. This noise mechanism is known as entropy noise, or indirect combustion noise. It generates both upstream-travelling acoustic waves, which can contribute to the feedback loop of combustion instabilities, and downstream-travelling acoustic waves, which can appear as exhaust noise… Show more

In gas turbines, unsteady combustion generates temperature fluctuations, which are often termed entropy waves. When these are accelerated through the turbine stages, they represent a significant source of noise. This noise mechanism is known as entropy noise, or indirect combustion noise. It generates both upstream-travelling acoustic waves, which can contribute to the feedback loop of combustion instabilities, and downstream-travelling acoustic waves, which can appear as exhaust noise. Prediction of entropy noise relies on (i) knowledge of how entropy waves are generated by unsteady combustion, (ii) how they advect with the turbulent flow through the combustion chamber , and (iii) how they generate acoustic waves upon reaching the turbine stages, where they are accelerated. As the strength of the two acoustic waves is directly proportional to the amplitude of the entropy wave reaching the inlet of the first stator blade, low order models need accurate knowledge of this wave strength. Many such models currently neglect the advection of the en-tropy wave through the turbulent combustion chamber, assuming that they advect undisturbed, except for a time delay. The aim of this work is to study the advection of entropy waves through a real combustion chamber flowfield, exhibiting realistic flow features such as swirl and recircula-tion zones. The combustor flowfield is found to weaken the strength of the entropy wave exiting the combustor chamber, but significant wave strength still remains at the inlet of turbine stages. Show less

A comprehensive list of publications can be found here, across many disciplines and research areas.