CFD Modeling of the Influence of Fuel Staging on the Mixing Quality and Flame Characteristics in a Lean Premixed Combustor

2010 ◽  
Author(s):  
Christina Schro¨dinger ◽  
Michael Oevermann ◽  
Oliver Kru¨ger ◽  
Arnaud Lacarelle ◽  
Christian O. Paschereit
Keyword(s):  
Experimental Data ◽  
Time Delays ◽  
Fuel Injection ◽  
Cfd Modeling ◽  
Reacting Flow ◽  
Fuel Distribution ◽  
Turbulent Schmidt Number ◽  
Mixing Quality ◽  
Fuel Staging ◽  
Flame Shape

In this paper, we investigate the feasibility and limitation of modeling non reacting and reacting flows of a premixed burner with steady RANS. The burner investigated here is a standard industrial swirl-inducing burner equipped with a staging of fuel injection. The simulation results on mixing quality, flame shape and position and convective time delays are compared to measurements which are performed in a water test rig and in a combustion chamber. The RANS simulations can qualitatively capture the trends observed from experimental data. The simulated mixing quality evolves in a similar way as the experimental data when the fuel distribution is changed. Using a turbulent Schmidt number of 0.2, the absolute values are in good agreement with the measured ones. Variations of the fuel injection distribution lead to changes in the flame shape and its stabilization location. The simulated reacting flow optimized with respect to the turbulent Schmidt/Prandtl number (Sct /Prt = 0.55) is able to predict the changes in flame shape and flame position. However, the shifting of the flame is not as distinct as observed in the experiments. This explains that variations in simulated convective time delays are also smaller than in reality. Nevertheless, the qualitative characteristics of the time delays depending on the fuel distribution parameter can be reproduced and absolute values are generally similar to those of the measurements.

Model Based Control of Emissions and Pulsations in a Premixed Combustor Using Fuel Staging

2009 ◽  
Author(s):  
Arnaud Lacarelle ◽  
Jonas P. Moeck ◽  
Christian O. Paschereit ◽  
Gregor Gelbert ◽  
Rudibert King
Keyword(s):  
Time Delays ◽  
Mixing Model ◽  
Extremum Seeking ◽  
Step Response ◽  
Nox Emissions ◽  
Pressure Pulsations ◽  
Model Based Control ◽  
One Dimensional ◽  
Mixing Quality ◽  
Fuel Staging

A mixing model of a swirl inducing premixed burner is derived from non-reacting investigation and used to control the fuel staging of the burner to ensure stability and low NOx emissions. The convective time delays, critical for the combustor stability, are obtained after identification of a step response of the outlet concentration with a one dimensional mixing model. The steady mixing is used to evaluate quantitatively the mixing quality which correlates with NOx emissions. Time delays as well as scalar unmixedness criteria derived from those measurements are used to predict the combustor stability and NOx emissions maps for different injection configurations at one operating point. The resulting model is used to extend an Extremum Seeking Controller, which adjusts the fuel repartition to reduce the pressure pulsations and NOx emissions.

On the Influence of Fuel Distribution on the Flame Structure of Bluff-Body Stabilized Flames

2013 ◽  
Vol 136 (4) ◽  
Author(s):  
Jeffery A. Lovett ◽  
Kareem Ahmed ◽  
Oleksandr Bibik ◽  
Andrew G. Smith ◽  
Eugene Lubarsky ◽  
...  
Keyword(s):  
Reaction Zone ◽  
Bluff Body ◽  
Fuel Injection ◽  
Flame Structure ◽  
Reacting Flow ◽  
Oh Radicals ◽  
Fuel Distribution ◽  
Jet In Crossflow ◽  
Wide Range ◽  
Stabilized Flames

This paper describes recent learning on the flame structure associated with bluff-body stabilized flames and the influence of the fuel distribution with nonpremixed, jet-in-crossflow fuel injection. Recent experimental and analytical results disclosing the flame structure are discussed in relation to classical combustion reaction zone regimes. Chemiluminescence and planar fluorescence imaging of OH* radicals as an indicator of the flame zone are analyzed from various tests conducted at Georgia Tech using a two-dimensional vane-type bluff-body with simple wall-orifice fuel injectors. The results described in this paper support the view that combustion occurs in separated flame zones aligned with the nonpremixed fuel distribution associated with the fuel jets that are very stable and contribute to flame stability at low fuel flow rates. The experimental data is also compared with computational reacting flow large-eddy simulations and interpreted in terms of the fundamental reaction zone regimes for premixed flames. For the conditions of the present experiment, the results indicate combustion occurs over a wide range of flame regimes including the broken reaction zone or separated flamelet regimes.

On the Influence of Fuel Distribution on the Flame Structure of Bluff-Body Stabilized Flames

2013 ◽  
Author(s):  
Jeffery A. Lovett ◽  
Kareem A. Ahmed ◽  
Oleksandr Bibik ◽  
Andrew G. Smith ◽  
Eugene Lubarsky ◽  
...  
Keyword(s):  
Reaction Zone ◽  
Bluff Body ◽  
Fuel Injection ◽  
Flame Structure ◽  
Reacting Flow ◽  
Oh Radicals ◽  
Fuel Distribution ◽  
Jet In Crossflow ◽  
Wide Range ◽  
Stabilized Flames

This paper describes recent learning on the flame structure associated with bluff-body stabilized flames and the influence of the fuel distribution with nonpremixed, jet-in-crossflow fuel injection. Recent experimental and analytical results disclosing the flame structure are discussed in relation to classical combustion reaction zone regimes. Chemiluminescence and planar fluorescence imaging of OH* radicals as an indicator of the flame zone are analyzed from various tests conducted at Georgia Tech using a two-dimensional vane-type bluffbody with simple wall-orifice fuel injectors. The results described in this paper support the view that combustion occurs in separated flame zones aligned with the non-premixed fuel distribution associated with the fuel jets that are very stable and contribute to flame stability at low fuel flow rates. The experimental data is also compared with computational reacting flow large-eddy simulations and interpreted in terms of the fundamental reaction zone regimes for premixed flames. For the conditions of the present experiment, the results indicate combustion occurs over a wide range of flame regimes including the broken reaction zone or separated flamelet regimes.

Numerical Investigation of the Flow Field and Mixing in a Swirl-Stabilized Burner With a Non-Swirling Axial Jet

2015 ◽  
Author(s):  
Tom Tanneberger ◽  
Thoralf G. Reichel ◽  
Oliver Krüger ◽  
Steffen Terhaar ◽  
Christian Oliver Paschereit
Keyword(s):  
Experimental Data ◽  
Flow Field ◽  
Swirling Flow ◽  
High Reactivity ◽  
Water Tunnel ◽  
Reacting Flow ◽  
Mean Flow ◽  
Lean Premixed Combustion ◽  
Mixing Quality ◽  
Flow Field Characteristics

In the present study numerical results of simulations, using RANS and LES, of the non-reacting flow in a swirl-stabilized burner are presented. The burner was developed for lean premixed combustion with high fuel flexibility at low emissions. An important challenge for a fuel-flexible, low emission combustor is the prevention of flashback for fuels of high reactivity, such as hydrogen, without compromising on lean blow out safety and mixing quality. Flashback safety can be increased by a sufficiently high and uniform axial velocity at the end of the mixing tube. In the investigated combustor the velocity deficit in the center of the mixing tube, which results from the swirl, is prevented by a non-swirling axial jet. In a parametric study the effect of different amounts of axial injection on the flow field is investigated. The results are validated with experimental data, gained from PIV measurements in a vertical water tunnel. It is shown that the mean flow field can be well captured by steady-state RANS simulations using a realizable k-ε turbulence model. The most suitable geometry is identified and, subsequently, transient LES simulations are conducted. The dynamic flow field characteristics are investigated. It was found that in spite of the high swirl, the flow field is quite stable and no dominating frequency is detected. The flow field of the swirling flow in the combustion chamber can be captured well using LES. Furthermore, the mixing quality is compared to the experiments, which are performed in a water tunnel. In contrast to the RANS simulation, the LES can qualitatively capture the spatial unmixedness observed from experimental data. All simulations were conducted using water as fluid.

Simulation of Laminar Liquid Jets in Gaseous Crossflow Before the Onset of Primary Breakup

2011 ◽  
Author(s):  
C.-L. Ng ◽  
K. A. Sallam
Keyword(s):  
Experimental Data ◽  
Fuel Injection ◽  
Liquid Jets ◽  
Liquid Jet ◽  
Jet Breakup ◽  
Primary Breakup ◽  
Density Ratios ◽  
First Time ◽  
Two Sides ◽  
Reduced Pressures

The deformation of laminar liquid jets in gaseous crossflow before the onset of primary breakup is studied motivated by its application to fuel injection in jet afterburners and agricultural sprays, among others. Three crossflow Weber numbers that represent three different liquid jet breakup regimes; column, bag, and shear breakup regimes, were studied at large liquid/gas density ratios and small Ohnesorge numbers. In each case the liquid jet was simulated from the jet exit and ended before the location where the experimental data indicated the onset of breakup. The results show that in column and bag breakup, the reduced pressures along the sides of the jet cause the liquid to move to the sides of the jet and enhance the jet deformation. In shear breakup, the flattened upwind surface pushes the liquid towards the two sides of the jet and causing the gaseous crossflow to separate near the edges of the liquid jet thus preventing further deformation before the onset of breakup. It was also found out that in shear breakup regime, the liquid phase velocity inside the liquid jet was large enough to cause onset of ligament formation along the jet side, which was not the case in the column and bag breakup regimes. In bag breakup, downwind surface waves were observed to grow along the sides of the liquid jet triggered a complimentary experimental study that confirmed the existence of those waves for the first time.

Dual-Location Fuel Injection Effects on Emissions and NO*/OH* Chemiluminescence in a High Intensity Combustor

2016 ◽  
Vol 138 (4) ◽  
Author(s):  
Ahmed O. Said ◽  
Ahmed E. E. Khalil ◽  
Ashwani K. Gupta
Keyword(s):  
Fuel Injection ◽  
Single Injection ◽  
Mixing Time ◽  
Fuel Mixture ◽  
Combustion Noise ◽  
Fuel Distribution ◽  
Mixture Preparation ◽  
Pollutants Emission ◽  
Fuel Mixing ◽  
Colorless Distributed Combustion

Colorless distributed combustion (CDC) has shown to provide ultra-low emissions of NO, CO, unburned hydrocarbons, and soot, with stable combustion without using any flame stabilizer. The benefits of CDC also include uniform thermal field in the entire combustion space and low combustion noise. One of the critical aspects in distributed combustion is fuel mixture preparation prior to mixture ignition. In an effort to improve fuel mixing and distribution, several schemes have been explored that includes premixed, nonpremixed, and partially premixed. In this paper, the effect of dual-location fuel injection is examined as opposed to single fuel injection into the combustor. Fuel distribution between different injection points was varied with the focus on reaction distribution and pollutants emission. The investigations were performed at different equivalence ratios (0.6–0.8), and the fuel distribution in each case was varied while maintaining constant overall thermal load. The results obtained with multi-injection of fuel using a model combustor showed lower emissions as compared to single injection of fuel using methane as the fuel under favorable fuel distribution condition. The NO emission from double injection as compared to single injection showed a reduction of 28%, 24%, and 13% at equivalence ratio of 0.6, 0.7, and 0.8, respectively. This is attributed to enhanced mixture preparation prior to the mixture ignition. OH* chemiluminescence intensity distribution within the combustor showed that under favorable fuel injection condition, the reaction zone shifted downstream, allowing for longer fuel mixing time prior to ignition. This longer mixing time resulted in better mixture preparation and lower emissions. The OH* chemiluminescence signals also revealed enhanced OH* distribution with fuel introduced through two injectors.

Flow Field and Hot Streak Migration Through a High Pressure Cooled Vanes With Representative Lean Burn Combustor Outflow

2018 ◽  
Vol 141 (4) ◽  
Author(s):  
Tommaso Bacci ◽  
Tommaso Lenzi ◽  
Alessio Picchi ◽  
Lorenzo Mazzei ◽  
Bruno Facchini
Keyword(s):  
Experimental Data ◽  
High Pressure ◽  
Flow Field ◽  
Fuel Injection ◽  
Leading Edge ◽  
Flame Stabilization ◽  
Lean Burn ◽  
Aero Engine ◽  
University Of Florence ◽  
Hot Streak

Modern lean burn aero-engine combustors make use of relevant swirl degrees for flame stabilization. Moreover, important temperature distortions are generated, in tangential and radial directions, due to discrete fuel injection and liner cooling flows respectively. At the same time, more efficient devices are employed for liner cooling and a less intense mixing with the mainstream occurs. As a result, aggressive swirl fields, high turbulence intensities, and strong hot streaks are achieved at the turbine inlet. In order to understand combustor-turbine flow field interactions, it is mandatory to collect reliable experimental data at representative flow conditions. While the separated effects of temperature, swirl, and turbulence on the first turbine stage have been widely investigated, reduced experimental data is available when it comes to consider all these factors together.In this perspective, an annular three-sector combustor simulator with fully cooled high pressure vanes has been designed and installed at the THT Lab of University of Florence. The test rig is equipped with three axial swirlers, effusion cooled liners, and six film cooled high pressure vanes passages, for a vortex-to-vane count ratio of 1:2. The relative clocking position between swirlers and vanes has been chosen in order to have the leading edge of the central NGV aligned with the central swirler. In order to generate representative conditions, a heated mainstream passes though the axial swirlers of the combustor simulator, while the effusion cooled liners are fed by air at ambient temperature. The resulting flow field exiting from the combustor simulator and approaching the cooled vane can be considered representative of a modern Lean Burn aero engine combustor with swirl angles above ±50 deg, turbulence intensities up to about 28% and maximum-to-minimum temperature ratio of about 1.25. With the final aim of investigating the hot streaks evolution through the cooled high pressure vane, the mean aerothermal field (temperature, pressure, and velocity fields) has been evaluated by means of a five-hole probe equipped with a thermocouple and traversed upstream and downstream of the NGV cascade.

scholarly journals A New Model Approach for Convective Wall Heat Losses in DQMOM-IEM Simulations for Turbulent Reactive Flows

2018 ◽  
Vol 141 (5) ◽  
Author(s):  
Yeshaswini Emmi ◽  
Andreas Fiolitakis ◽  
Manfred Aigner ◽  
Franklin Genin ◽  
Khawar Syed
Keyword(s):  
Experimental Data ◽  
Heat Losses ◽  
Reacting Flow ◽  
Jet Flame ◽  
New Model ◽  
Turbulent Reactive Flows ◽  
Novel Method ◽  
The One ◽  
The Impact ◽  
Model Approach

A new model approach is presented in this work for including convective wall heat losses in the direct quadrature method of moments (DQMoM) approach, which is used here to solve the transport equation of the one-point, one-time joint thermochemical probability density function (PDF). This is of particular interest in the context of designing industrial combustors, where wall heat losses play a crucial role. In the present work, the novel method is derived for the first time and validated against experimental data for the thermal entrance region of a pipe. The impact of varying model-specific boundary conditions is analyzed. It is then used to simulate the turbulent reacting flow of a confined methane jet flame. The simulations are carried out using the DLR in-house computational fluid dynamics code THETA. It is found that the DQMoM approach presented here agrees well with the experimental data and ratifies the use of the new convective wall heat losses model.

Assessment of CFD Approaches for Next-Generation Combustor Design

2014 ◽  
Author(s):  
Kumud Ajmani ◽  
Hukam C. Mongia ◽  
Phil Lee
Keyword(s):  
Experimental Data ◽  
Direct Injection ◽  
Operating Conditions ◽  
Reacting Flow ◽  
Cfd Analysis ◽  
Next Generation ◽  
Lean Direct Injection ◽  
Liquid Spray ◽  
Exit Temperature ◽  
Computational Predictions

An effort was undertaken to perform CFD analysis of fluid flow in Lean-Direct Injection (LDI) combustors with axial swirl-venturi elements for next-generation LDI-2 design. The National Combustion Code (NCC) developed at NASA Glenn Research Center was used to perform reacting flow computations on an LDI-2 combustor configuration with thirteen injector elements arranged in four fuel stages. Reacting computations were performed with a consistent approach for mesh-optimization, liquid spray modeling and kinetics modeling. Computational predictions of Emissions Index (EINOx) and combustor exit temperature were compared with two sets of experimental data at medium and high-power operating conditions, for two different pressure-drop conditions in the combustor. The NCC simulations predicted the combustor exit temperature to within 1–2% of experimental data. The accuracy of the EINOx predictions from the NCC simulations was within 10% to 30% of experimental data.

Sign in / Sign up

Export Citation Format

Share Document