Search

Flame particles (FP) are massless, virtual particles which follow material points on the flame surface. This work presents a tracking algorithm for FPs which utilizes barycentric coordinates. The methodology can be used with any cell shape in the computational mesh and allows computationally fast spatial interpolation as well as efficient determination of the intersection of FP trajectories with iso-surfaces. In contrast to previous flame particle tracking (FPT) approaches, the code is fully parallelized and can therefore be used in-situ during the simulation. It also includes fully parallelized computation of flame consumption speed by integrating reaction rates along a line normal to the flame surface at each FP position. Direct numerical simulations of laminar pulsating premixed hydrogen–air Bunsen flames serve as validation cases and showcase the added value of tracking material points for studying local flame dynamics. Exciting the inlet flow harmonically with frequencies equal to the inverse flame time scale leads to a pulsating mode where the flame front is corrugated. Ten times higher frequencies nearly resemble the steady state solution. The FPs are seeded along the flame surface and are used to track the unsteady diffusive, convective and chemical contributions at arbitrary points on the flame front over time. Their trajectories reveal a phase shift between the unsteady flame stretch rate and local flame speed of the order of 0.1 flame time scales for rich hydrogen flames. This is caused by a time delay between straining and stretch due to curvature. The reason is that diffusive processes follow the time signal of curvature while chemical processes are most strongly affected by the straining rate, which dominates the high Lewis number hydrogen flames investigated. This time history effect may help to explain the large scattering in the correlation of local flame speed with flame stretch found in turbulent flames.

Large eddy simulations (LES) have been carried out to calculate turbulent flame propagation in a fan-stirred combustion bomb and a Bunsen-type burner. Objective of the work is to reveal the main mechanism of increased flame wrinkling due to elevated pressure and to assess the ability of the turbulent flame-speed closure (TFC-class) combustion model to reproduce the enhancement of flame wrinkling or burning rate at elevated pressures. The simulations have been performed for a premixed methane/air mixture at equivalence ratio 0.9 and the pressure has been varied from 1 bar to 5 bar. The turbulent kinetic energy is found to increase with pressure in the high frequency range, indicating reinforced small-scale turbulent fluctuations at elevated pressure. The reason is attributed to the increased turbulent Reynolds number with pressure, which shifts the turbulent energy spectra to the higher wave number range. A reinforced flame wrinkling and an increased total burning rate are obtained at elevated pressure, which is in accordance with results from previous high-pressure combustion experiments. In addition, applying the same method to a quiescent flow in the bomb vessel reveals a decrease of the overall burning rate with pressure, which agrees with the behaviour of laminar flame speed at elevated pressures. Therefore, the beneficial effect of increased burning rate or flame wrinkling at elevated pressure can be explained by the enhanced small-scale turbulent fluctuations along with formation of small-scale vortices and flame structures, which over-compensate the reduced local laminar burning velocity at high pressures. The calculated amplification rates of flame wrinkling factor at increased pressures show a reasonable agreement with measured data for both fan-stirred bomb and Bunsen flame configurations, without using any additional adjusting parameters for considering the pressure effect. The results justify the applicability of the current TFC-LES method for high-pressure combustion processes.

Numerical simulations have been conducted for a novel double-concentric swirl burner, which is specifically designed for combustion of sulfur with a high power density. The burner serves as a major component of an enclosed conversion cycle, which uses elemental sulfur as a carbon-free chemical energy carrier for storing solar energy. The focus of the work is to assess operability of the burner and NO

This work presents a numerical study of the initial turbulent flow characteristics prior to ignition in a fan-stirred combustion vessel. The moving mesh methodology has been applied to account for ...

Turbulent swirl flames using fuel sprays are found in many combustion systems. Since sprays with different characteristics can be produced, pulsed spray injection allows the realization of differen...

As an alternative to usual combustion processes with air as oxidant, oxy-fuel or oxygen-enhanced processes are attractive in high temperature thermal or thermochemical processes, novel power plant concepts or in gasification processes. Especially for production of basic chemicals or synthetic fuels, high-purity synthetic gas without nitrogen dilution is required for further industrial processing. In case of oxy-fuel combustion, the absence of nitrogen leads to higher flame temperatures and larger concentration of major species as well as intermediate species. Because of the increased reactivity, most combustors are designed in non-premixed configuration. In the present work, the extinction limits of non-premixed CH4/N2/O2 flames were numerically calculated using the CHEMKIN Pro opposed flow code together with the GRI 3.0 mechanism. The inlet compositions were varied from high nitrogen dilution to pure oxy-fuel conditions. Additionally, the stoichiometric mixture fraction was varied by shifting the (inert) N2-fraction stepwise from the oxidizer to the fuel side. Therewith, the equilibrium temperature of the mixture is constant for a fixed N2-dilution. The results show, that as expected, the extinction occurs at higher strain rates with increasing the flame temperatures by reducing the N2-diluation. Additionally, the extinction limits strongly increase for a fixed N2-dilution with increasing the stoichiometric mixture fraction. The shift of the N2 from the oxidizer to the fuel side leads to a movement of the stagnation plane towards the fuel side and hence, a significant change of the temperature field and the flame structure. Thus, the radical chain branching occurs at higher temperature level, which strengthen the flame resistance against strain. The increase of the stoichiometric mixture leads to a shift of the flame towards the fuel side and crosses at high values the stagnation plane.

In this work, a hybrid Euler-Lagrangian solver for dense spray systems is developed specifically for cases where film creation by accumulation of liquid droplets at the walls plays a crucial role. EulerLagrangian solvers are commonly used to describe the spray with predefined spray characteristics. The Lagrangian particles represent liquid drops moving along the continuous gaseous phase. This approach assumes a small particle size compared to the cell size and it is unable to capture the breakup behavior of liquid jets in the presence of instabilities. VOF methods, on the other hand, are not a computationally feasible option when it comes to small droplet sizes as a result of liquid atomization because they have to be fully resolved by the computational mesh. Hence, multiscale simulations are required to bridge the gap between the two methods combining subgrid droplets in Lagrangian approaches and large liquid structures in VOF methods. In the present work, a multiscale approach is developed where Lagrangian particles representing small droplets are tracked through the continuous phase until they hit a wall or a liquid-gas interface represented by a continuous VOF field. At the time of impact, the Lagrangian particles are removed and the mass and momentum of these particles are transferred to the VOF field. This allows having a spray consisting of subgrid droplets computed with a Lagrangian particle tracking (LPT) approach and liquid films at the walls with VOF method. The method represents a one-way coupling, converting Lagrangian particles to Eulerian liquid phase (VOF) and has been implemented into the open-source CFD code OpenFOAM. Subsequently, the solver has been tested in different scenarios to ensure mass and momentum conservation. Positive test results encouraged its use to gain insight on the fluid flow in a real cylindrical compression-dissolution unit where the liquid is sprayed from the top while simultaneously the gas is compressed from the bottom. Dynamic mesh technique is used to account for the compression with a moving piston.

To meet future climate goals, the efficiency of combustion devices has to be increased. This requires a better understanding of the underlying physics. The simulation of turbulent flames is a challenge because of the multi-scale nature of combustion processes: relevant length scales span over five orders of magnitude and time scales over more than ten. Because of this, the direct numerical simulation (DNS) of turbulent flames is only possible on large supercomputers. A DNS solver for chemically reacting flows implemented in the open-source framework OpenFOAM is presented. The thermo-chemical library Cantera is used to compute detailed transport coefficients based on kinetic gas theory. The multi-scale nature of time scales, which are much lower for the combustion chemistry than for the flow, are bridged by an operator splitting approach, which employs the open-source solver Sundials to integrate chemical reaction rates. Because the direct simulation of turbulent flames has to be performed on supercomputers, special care has been taken to improve the computational performance. A tool was developed which generates highly optimized C++ source code for the computation of chemical reaction rates. Additionally, a load balancing approach specifically made for the computation of chemical reaction rates is employed. In total, these optimizations can reduce total simulation times by up to 70 %. The accuracy of the new solver is assessed from different canonical testcases: 2D and 3D Taylor-Green vortex simulations show that the solver can reach up to fourth order convergence rates and that results differ by less than 1 % when compared to spectral DNS codes. Molecular diffusion and chemical reaction rates are compared to solutions of 1D flames from Cantera, showing perfect agreement. The solver is used to simulate the Sydney/Sandia burner. The simulation is performed on one of Germany's largest supercomputer on 28 800 CPU cores, employing 150 million cells and a chemical reaction mechanism with 19 species and about 200 reactions. Comparison with experimental data shows excellent agreement for time averaged and RMS values.

The article Quasi-DNS Dataset of a Piloted Flame with Inhomogeneous Inlet Conditions written by Thorsten Zirwes, Feichi Zhang, Peter Habisreuther, Maximilian Hansinger, Henning Bockhorn, Michael Pfitzner, Dimosthenis Trimis, was originally published Online First without Open Access.

To ensure compliance with stricter regulations on exhaust gas emissions, new industrial burner concepts are being investigated. One of these concepts is the matrix burner, consisting of an array of premixed, non-swirling jet flames. For the design of such burners, the prediction of fundamental burner properties is mandatory. One of these essential quantities is the lean blowout limit (LBO), which has already been investigated experimentally. This study investigates the possibility of numerical LBO prediction using a tabulated chemistry approach in combination with Large-Eddy-Simulation turbulence modeling. In contrast to conventional swirl burners, the numerical description of blowout events of multi jet flames has not yet been studied in detail. Lean blowout simulations have therefore been conducted for multiple nozzle variants, varying in their diameter and global dump ratio for a variety of operating conditions, showing their general applicability. A procedure to induce LBO is introduced where a stepwise increase in total mass flow is applied. LBO is determined based on the temporal progress of the mean reaction rate. A comparison with measurements shows good agreement and demonstrates that the procedure developed here is an efficient way to predict LBO values. Further investigations focused on the flame behavior when approaching LBO. The flame shape shows a drastic change from single jet flames (stable conditions) to a joint conical flame approaching LBO, which increases in length for increasing inlet velocity, showing the importance of jet interaction at LBO.

Identifying combustion regimes in terms of premixed and non-premixed characteristics is an important task for understanding combustion phenomena and the structure of flames. A quasi-DNS database of the compositionally inhomogeneous partially premixed Sydney/Sandia flame in configuration FJ-5GP-Lr75-57 is used to directly compare different types of flame regime markers from literature. In the simulation of the flame, detailed chemistry and diffusion models are utilized and no turbulence and combustion models are used as the flame front and flow are fully resolved near the nozzle. This allows evaluating the regime markers as a post-processing step without modeling assumptions and directly comparing regime markers based on gradient alignment, drift term analysis and gradient free regime identification. The goal is not to find the correct regime marker, which might be impossible due to the different set of assumptions of every marker and the generally vague definition of the partially premixed regime itself, but to compare their behavior when applied to a resolved turbulent flame with partially premixed characteristics.

OpenFOAM is one of the most popular open source tools for CFD simulations of engineering applications. It is therefore also often used on supercomputers to perform large eddy simulations or even direct numerical simulations of complex cases. In this work, general guidelines for improving OpenFOAM’s performance on HPC clusters are given. A comparison of the serial performance for different compilers shows that the Intel compiler generally generates the fastest executables for different standard applications. More aggressive compiler optimization options beyond O3 yield performance increases of about 5 % for simple cases and can lead to improvements of up to 25 % for reactive flow cases. Link-time optimization does not lead to a performance gain. The parallel scaling behavior of reactive flow solvers shows an optimum at 5000 cells per MPI rank in the tested cases, where caching effects counterbalance communication overhead, leading to super linear scaling. In addition, two self-developed means of improving performance are presented: the first one targets OpenFOAM’s most accurate discretization scheme “cubic”. In this scheme, some polynomials are unnecessarily reevaluated during the simulation. A simple change in the code can reuse the results and achieve performance gains of about 5 %. Secondly, the performance of reactive flow solvers is investigated with Score-P/Vampir and load imbalances due to the computation of the chemical reaction rates are identified. A dynamic-adaptive load balancing approach has been implemented for OpenFOAM’s reacting flow solvers which can decrease computation times by 40 % and increases the utilization of the HPC hardware. This load balancing approach utilizes the special feature of the reaction rate computation, that no information of neighboring cells are required, allowing to implement the load balancing efficiently.

This work assesses the effect of using different closure concepts for the spatially filtered mean reaction rate on the resolved spectral response of turbulent heat release in large eddy sim...

As an alternative to usual combustion processes with air as oxidant, oxy-fuel processes are attractive in high temperature thermal or thermochemical processes, novel power plant concepts or in gasification processes. Especially for production of basic chemicals or synthetic fuels, high-purity synthetic gas without nitrogen dilution is required for further industrial processing. In case of oxy-fuel combustion, the absence of nitrogen leads to higher flame temperatures and larger concentration of major species as well as intermediate species. In the present work, freely propagating methane-oxygen-nitrogen flames were numerically calculated using a 1D model from lean to rich conditions in order to investigate the appearance of super-adiabatic flame temperatures (SAFT). The calculations were performed for equivalence ratios of 0.5 The SAFT phenomenon was identified based on temperature and species profiles. Additionally, a detailed analysis of convection, diffusion and chemical source of both, temperature and species, were performed in order to investigate the role of physical transport processes on SAFT. The results showed that the maximum flame temperature exceeds the equilibrium temperature for equivalence ratios Φ > 0.9. Two different regimes were identified, where SAFT phenomenon appears. The first regime was found in slightly rich conditions (1.0 2.0). A first maximum of temperature difference is observed at an equivalence ratio of Φ = 1.5. Here, an exaggeration of approximately 120–180 K, depending on the applied reaction mechanism, was found for standard conditions. The first maximum at Φ = 1.5 correlates with the maximum concentration of the H-radical, which plays a key role in the first SAFT regime. A minimum temperature difference of 50 K was identified at an equivalence ratio of Φ = 2.1. While increasing the equivalence ratio further, the maximum flame temperature exceeded the equilibrium up to almost 400 K at Φ = 3.0 in the second SAFT regime. An increased preheating temperature enhanced the occurrence of SAFT in the first regime and degraded it in the second regime. Elevated pressure leads to the opposite effects with decreased SAFT in the first and increased SAFT in the second regime.

Gas combustion occurring within porous inert media (PIM) results in intensified heat and mass transfer in comparison to free laminar combustion. This leads to considerably lower emissions of pollutant species, that is, NOx and CO, larger flame stability range, lower thermo-acoustic instabilities, gradual response to operative changes, and geometric flexibility of design. In this work, experiments were carried out to describe the influence of flow rate, pressure, temperature, and air-to-fuel ratio on the flame stability within PIM at nearly adiabatic conditions. The macroscopic thermal flame thickness and the burning velocity were calculated from measured temperature profiles and flow rates. As the main result, it was observed that the pressure effect on the burning velocity can differ considerably from that of laminar free flames, that is, the increase of pressure can positively affect the burning velocity. The obtained results were supported by theoretical modeling. A new model to average the species production rates in a volume-averaged numerical model was proposed. The model considered spatial deviations of temperature and mass fractions along the cross-sectional area using presumed probability density functions. Using this new model, results of burning velocities and profiles of average temperature were satisfactorily predicted when compared to experimental results shown here. Furthermore, easy-to-use tools for porous burner design were formulated and validated.

A quasi-DNS of the partially premixed turbulent Sydney flame in configuration FJ200-5GP-Lr75-57 has been conducted using detailed molecular diffusion for multi-component mixtures and complex reaction mechanisms. In order to study flame dynamics like regime transition in this flame for the development of new combustion models and to directly compare the quasi-DNS to different LES models, the simulation results are compiled into a data base. Because the simulation was performed with OpenFOAM, we demonstrate the quasi-DNS capabilities of OpenFOAM by performing canonical test cases. They attest that OpenFOAM’s cubic discretization has lower numerical diffusion compared to classical central difference schemes and can reach higher than second order convergence rate in some cases. The quasi-DNS of the Sydney flame is conducted with a self-developed reacting flow solver which is able to accurately compute molecular diffusion coefficients from kinetic gas theory and employs a fast implementation for detailed reaction mechanisms. The computational mesh is shown to be able to resolve the flow as well as the flame front sufficiently for the quasi-DNS. Comparisons with experimental data also show that the simulation can quantitatively reproduce measured time-mean and time-RMS statistics.

A burner system for the efficient and clean combustion of sulfur is introduced, which serves as a key component in a novel solar power cycle using sulfur as chemical storage medium of solar energy. In order to validate the proposed design concept, highly-resolved numerical simulations have been performed. The current setup is operated with a thermal load of 20 kW or power density of 5 MW/m3. Two nozzle configurations with different swirl intensities (SI) of the airflow are studied. A large inner recirculation zone is observed for the nozzle with a high SI (HSI), which leads to a strong radial dispersion of the sulfur spray and a broad, short flame in the combustion chamber; although this HSI design is beneficial from the viewpoint of flame stabilization, it causes a large number of sulfur droplets hitting the chamber wall. In contrast, the nozzle design with a low SI (LSI) yields a narrow spray and a long jet flame, with much less droplets hitting the wall. The HSI nozzle shows an overall higher flame temperature compared with the LSI nozzle, which is confirmed to be caused by burning at a higher local fuel equivalence ratio. This is attributed to the strong inner recirculation flow generated by the high swirl intensity, which results in an enhanced evaporation and mixing of sulfur droplets with air. In terms of operability and NOx emission, the LSI burner is preferred due to less sulfur droplets hitting the chamber wall and the lower flame temperature.

Highly resolved numerical simulations have been conducted for a generic, coaxial air-blast atomizer designed for fundamental research of entrained flow gasification processes. Objective of the work is to gain a detailed knowledge of the influence of elevated reactor pressure on the primary atomization behaviour of high-viscous liquid jets. In agreement with measured breakup morphology and breakup regimes proposed in literature, the simulations yield a pulsating mode instability of liquid jet, along with disintegrations of fibre-type liquid fragments for different pressures. From the mechanism point of view, the breakup process has been shown to be triggered by concentric, axisymmetric ring vortices, which disturb the liquid jet surface in a first stage and penetrate further into the intact core, leading to interfacial instabilities and pinch-off of liquid ligaments. The liquid jet breaks up faster at elevated pressure, leading to a shorter core length L C . The calculated exponent ( b ≈ − 0 . 5 ) of the power law for fitting the decrease of L C with p agrees well with measured correlations from literature in terms of varied momentum flux ratio M and Weber number W e G , although water jets, atmospheric pressure and different air-assisted, external mixing nozzles were used in these works. Therefore, the effect of elevated pressure is equivalent to that of increased M or W e G , which scale linearly with p or the gas density for the current setup. The specific kinetic energy of liquid k L has been found to be increased with p , which is particularly pronounced in the high frequency range. A first-order estimate has been proposed, which can be used for the evaluation of liquid kinetic energy or droplet velocity within the spray. The results have been validated by simulations with twice-refined resolution, yielding a grid-independence behaviour with respect to the primary breakup characteristics. However, the follow-up processes with secondary breakup and spray dispersion are reproduced better by using the finer grid.

In order to extend the operation regime of existing gas turbine combustion systems to lower the minimum loads, the applicability of matrix burners (arrays of jet flames) as an alternative to conventional swirl stabilized burners has been considered. In comparison to well-studied single jet flame systems, the effects of geometry and thermodynamic parameters on characteristics of matrix burner systems have not been studied in detail. Information, which is essential for design processes e.g. scaling of matrix burners, is not yet available in public domain. This work involves a systematic investigation of a matrix burner system operating at highly turbulent flow conditions (Reynolds Number ≈ 20000) prevailing in gas turbine combustion systems. In order to understand the effects of geometrical scaling, three variants of jet diameter have been investigated. A detailed test campaign including lean blow out limits detection, velocity field measurement and hydroxyl radical (OH*) chemiluminescence recording has been conducted. Influence of variation in stoichiometry and exit velocity of fuel-air mixture has been captured. The results show that it is possible to generalize the scaling of the matrix burner using the well-known Peclet criterion.

Combustion is an important part of most current and future overall energy-conversion systems, especially if using renewable fuels in energy-storage concepts. Therefore, the laminar flame speed, which is a key parameter for the design of combustion systems, needs to be known for a growing multitude of different thermodynamic conditions and fuels. The spherically expanding flame method is one of the few techniques that enables the flame speed to be measured under particular conditions such as elevated pressure and temperature as well as under turbulent conditions, which are important for energy-conversion applications. The radius of a spherically propagating flame is tracked and used for evaluation of the flame speed. Usually, the flame is assumed to be infinitely thin. To assess the influence of this assumption, direct numerical simulations were conducted for the experimental setup and compared with measurements and correlations from the literature. The flame speed determined by the consumption rate of fuel, which takes a finite thickness of the flame into account, was found to be always larger than the flame speed computed by assuming an infinitely thin flame. The difference between these flame speeds was observed to be as large as approximately 10–20 % in the evaluation range of the measured flame radii, which decreases with growing flame radius. This gives rise to the discrepancies in the flame speeds obtained from different measurement methods. An analytical estimation for this difference was developed as a function of the flame radius, which showed quantitatively good agreement with the simulation results and may be used for experimental validations of the flame speed. Both premixed H2/air and CH4/air flames with equivalence ratios ranging from lean to rich conditions were studied.

A first-order estimation for the prediction of combustion-generated noise from turbulent premixed flames has been explored. The method is based on Lighthill's acoustic analogy and uses measurement data from high-speed imaging of chemiluminescent emissions as input. It has been shown by both 1D and 3D detailed numerical simulations that the local generation of OH* correlates strongly with the heat released from the chemical reaction, especially in the fuel-lean range. As the chemiluminescence measurement gathers light only along the viewing direction, the line-of-sight summed values of heat release rate and OH* concentration have been evaluated from the 3D simulation and a quasi-linear relationship has been identified for these integral values. Hence, a proportionality relation has been applied for computation of the integral heat release from measured chemiluminescence intensity. An analytical solution of Lighthill's wave equation serves as a transfer function, which takes fluctuations of the total heat release rate or light intensity as input to calculate the sound radiation in the far field. This approach has been applied to a generic burner operated with a premixed methane/air jet flame. Good quantitative agreement is obtained between the sound pressures derived from the chemiluminescence measurement and microphone data, which highlights the potential of the method for applications to more complex flame configurations.

Numerical simulations employing detailed reaction kinetics and transport have been applied to calculate steady-state and harmonically excited laminar hydrogen–air flames with equivalence ratios ranging from fuel-lean to fuel-rich mixtures. The objective is to gain a basic understanding of the influence of unsteady flow stretching on the flame’s response. The flame displacement speed and consumption speed have been evaluated, together with stretch, curvature and strain. In the steady-state simulations, a quasi-linear correlation has been obtained for the flame speeds with stretch for a moderate range, which confirms the applicability of the asymptotic relation for small stretch. In the case of unsteady flow conditions, a phase shift is observed between the oscillations of the fluid flow and the flame surface position, indicating a delayed response of the flame to changing flow conditions. The corresponding delay time or relaxation time is found to be depending on the turnover time of the flow oscillations. The flame response in terms of displacement from its steady state location is shown to be dampened at higher excitation frequencies and for flames with large flame transit times τL0. Due to the delayed response, the local consumption speeds at different phase angles show different dependences on flame stretch. This discrepancy is large for flames in flows excited with lower frequencies as well as equivalence ratios with smaller flame transit times. Also, a larger variation of consumption speeds from different phase angles over flame stretch has been observed for regions with positive stretch, because the internal structure of the flame is altered, leading locally to smaller flame transit times. This explains the large scattering of flame propagation statistics in turbulent combustion, where the flow is characterized by a cascade of interacting vortices with different turnover frequencies. The unsteady effect is quantitatively analysed by comparing the characteristic times of the flame and the flow. In this way, an extension of the Markstein regression has been provided based on the simulation results, giving Ma as a function of the Damkohler number defined by Da = τ f l o w / τ L 0 = S L 0 / δ L 0 / f . Evaluating the consumption speed for different phases during a turnover period of the oscillations and averaging over the period results in a correlation of the averaged Markstein number with the Damkohler number.

This study investigates the ignition/combustion of di-tertiary-butyl peroxide (DTBP) over a wide range of fuel/air ratios using numerical flame calculations to gain insight into species profiles and ignition/combustion characteristics of DTBP/air mixtures. The results are flame zone and reaction zone structures as well as the corresponding laminar flame speeds and propagation speeds of reaction zones. Several detailed reaction mechanisms are evaluated in the flame simulations, and sensitivity of the computed results to important reaction steps was evaluated. The ignition of DTBP in the absence of oxygen belongs in the category of thermal runaway because there is no chain branching associated with the early stages of the reaction; i.e., decomposition steps lead to acetone and ethane. In the presence of oxygen, the development of a flame is more complex and exhibits a two-stage ignition, where the first stage, at lower temperatures, involves the thermal decomposition of DTBP to acetone, which further oxidiz...

The conversion of low-grade fossil and biogenic energy resources (petcoke, biomass) to a synthesis gas in a high pressure entrained flow gasification process opens a wide spectrum for high efficient energy conversion processes. The synthesis gas can be used for production of methane (SNG), liquid fuels (BtL, CtL) or as fuel for operation of a gas turbine in a combined cycle power plant (IGCC). The production of a tar free high quality syngas is a challenging objective especially due to the fact that typical liquid or suspension fuels for entrained flow gasifiers feature viscosities up to 1000 mPas. Fuel droplet conversion at typical entrained flow gasification conditions is characterized by heat up, evaporation and subsequent degradation of the vapour phase. To guarantee a high fuel conversion rate in the gasifier an efficient atomization of the fuel is required. Mainly twin-fluid burner nozzles are used for atomization of those typically high viscous fuels. The present study is focused on the assessment of the accuracy of CFD computations for the primary breakup of high-viscosity liquids using an external mixing twin fluid nozzle. In a first step experiments were performed with a Newtonian glycerol-water-mixture featuring a liquid viscosity of 400 mPas. Jet breakup was investigated using a high speed camera as well as PIV and LDA-System for a detailed investigation of the flow field. In a second step the experimental results serve as reference data to assess the accuracy of CFD computations. Compressible large eddy simulations (LES) were performed to capture the morphology of the primary breakup as well as the important flow field characteristics. A Volume of Fluid (VOF) approach was used to track the unsteady evolution and breakup of the liquid jet. Comparison of experimental and numerical results showed good agreement with respect to breakup frequency, velocity fields and morphology.

New values for the viscous and inertial coefficients of the Ergun equation to calculate the pressure drop within open cell ceramic sponges made of alumina and silicon infiltrated silicon carbide (SiSiC) are provided based on the direct pore level simulations (DPLS) results. The fluid flow through ceramic structures of different cell densities and porosities are simulated using standard Navier–Stokes equation without any simplification. The computational grid data are reconstructed from three-dimensional computer tomography scans of different reticulated porous media. Influence of tortuosity on the pressure drop is included in a Ergun equation based pressure drop correlation. The new correlation is dependent on the sponge properties such as, cell diameter, specific surface area and flow tortuosity.

The computation of chemical reaction rates is commonly the performance bottleneck in CFD simulations of turbulent combustion with detailed chemistry. Therefore, an optimization method is used where C++ source code is automatically generated for arbitrary reaction mechanisms. The generated code is highly optimized for the chosen mechanism and contains all routines for computing chemical reaction rates. In this work, the serial performance of the automatically generated source code, which in an earlier work only used ISO C++, is further improved by utilizing two compiler extensions: restrict and __builtin_assume_aligned. Introducing these two extensions to the generated code reduces the time for computing chemical reaction rates by up to 50% for the investigated reaction mechanism and total simulation time by up to 25%. Compared to OpenFOAM’s standard chemistry implementation, the new code is faster by a factor of 10. This work discusses the effect of the two compiler extensions on performance by looking at two specific kernel functions from the automatically generated code and the effect on the assembly generated by the gcc and Intel compilers. The newly optimized code is used to evaluate the performance gain in a large scale parallel case, which simulates an experimentally investigated turbulent flame of laboratory scale on 14,400 CPU cores on the Hazel Hen cluster at HLRS. In the simulation, no combustion models are used and the flame is resolved down to the smallest length scales. With this approach, comparison of measured data with the simulation shows very good agreement. Using the optimized code including compiler extensions, total simulation time decreases by 20% compared to the same code without compiler extensions. A comprehensive database from the simulation results has been assembled and will consist of 10 TB of 3D and 2D transient field variables.

Detail investigations on the primary breakup of high-viscosity liquids using external-mixing twin-fluid nozzles at increased system pressure are scarce. Therefore, the research work of the present study is focused on the investigation of pressure influence (1 - 11 bar (abs)) on the primary breakup by numerical simulation based on a previously studied nozzle [Müller et al., ASME Turbo Expo 2016, GT2016-56371]. The pressure influence was investigated for two liquids applying a wide range of viscosities (100 mPa s; 400 mPa s) and two atomizing air velocities (58 m/s; 74 m/s). To describe the disintegration process of the fluids, characteristic features like liquid jet morphology, breakup length and breakup frequency were evaluated. The primary breakup was investigated using the open source CFD software OpenFOAM. To gather the morphology of the primary breakup and the flow field characteristics compressible large eddy simulations (LES) were performed and the movement of the gas-liquid interface was captured by means of the Volume of Fluid-Method (VOF). The conducted simulations showed good agreement with experimental results with respect to the characteristic features (e.g. morphology and breakup length) and revealed a decrease of the breakup length with increasing ambient pressure for a constant liquid mass flow and atomizing air velocity. Moreover, those findings will contribute to a better understanding of the physics of the breakup of high-viscosity liquid jets and as well to create an experimentally validated CFD based tool for future burner development and optimization.

Gas turbine combustion engineers strive to obtain high flame stability at low power conditions (idle); whereas good emissions characteristics are desired at high power conditions. Although lean combustors provide good emissions characteristics, they suffer from the issue of poor flame stability. Often a pilot flame along with a lean main flame is used in order to improve flame stability issue. Positioning of an axial jet concentric to a swirl jet is one of commonly used configurations for a pilot flame. The interaction between both co-annular jets leads to distinct flow patterns in the combustor and subsequently the flame structure is impacted. If the axial jet is able to penetrate the inner recirculation zone (IRZ) generated by the swirled flow of the main jet, then a jet type of flame is obtained. The jet flame in vicinity of co-annular swirled flow is also denoted as type 1 flame, and is known to have good flame stability. On the other hand, if the pilot jet is not able to penetrate the IRZ, then the recirculating type of flame is obtained. A recirculating flame type has good emissions characteristics but suffers from poor flame stability. In this work the numerical predictions have been performed to gain more insight into occurrences of these two different flame structures which have been experimentally recorded. Using ANSYS Fluent CFD software, non-reactive steady state turbulent flow simulations have been performed to understand the flow and mixing field in a 3D combustor. Laser Doppler Anemometry (LDA) and hydroxyl radical (OH*) chemiluminescence have been used to measure the flow and characterize the flame structure, respectively.

The atomization of liquid fuel in gas turbines is one of the key factors that affect the combustion performance. After the adaptation of the k-ϵ-turbulence model for two-phase flows and the definition of the film thickness, this work investigates and reports the influence of different parameters on the behavior of the liquid film on the prefilmer of airblast atomizers. A correlation that can be used to predict the mean film thickness has been developed and is proposed in the present study. This correlation can be used in order to optimize the design of airblast atomizers.

In the case of turbine combustors operating with liquid fuel the combustion process is governed by the liquid fuel atomization and its dispersion in the combustion chamber. By highly unsteady flow field conditions the transient interaction between the liquid and the gaseous phase is of interest, because it results in a temporal variation of air–fuel ratio which leads to a fluctuating temperature distribution. The objective of this research was the investigation of transient flow field phenomena (e.g. large coherent structures) on droplet dynamics and dispersion of an isothermal flow (of inert water droplets) as a necessary first step towards a full analysis of spray combustion in real-life devices. The advanced injector system for lean jet engine combustors PERM (Partial Evaporated Rapid Mixing) was applied, generating a dilute polydispersed spray in a swirled flow field. Experiments were performed using Phase Doppler Anemometry (PDA) and a patternator to determine the droplet polydispersity, concentration maps, and velocity profiles in the flow. An important finding is the effect of large-scale coherent structures due mainly to the precessing of the vortex core (PVC) of the swirling air jet on the particle dispersion patterns. The experimental results then serve as reference data to assess the accuracy of the Eulerian–Lagrangian computations using a Large Eddy Simulation (LES), a Unsteady Reynolds-Average Navier–Stokes Simulation (URANS) and two simplified (steady-state) simulations. There, a simplified droplet injection model was used and the required boundary conditions of injected droplet sizes were obtained from measurements. Important transient effects of deterministic droplet separation observed during experiments, could be perfectly replicated with this injection model. It is convincingly shown, through extensive computations, that the resolution of instantaneous vortical structures is indeed crucial; hence the LES, or a reasonably-well resolved URANS are preferred over the steady-state solutions with additional, stochastic-type, turbulent dispersion models.

This work is a comprehensive study of combustion in porous inert media (PIM) with focus on the effects of pressure and air/fuel equivalence ratio. Experiments on lean flame stability limits, flame stabilization and temperature profiles in porous inert media burners of different geometry have been carried out at nearly adiabatic conditions. The experiments were complemented by numerical simulations of combustion in porous inert media, employing a volume-averaged 1D model and 3D direct pore level simulation (3D DPLS) on real geometries of sponge like structures. A definition of the burning velocity is proposed to compare the geometrically different burners investigated. The numerically obtained macroscopic thermal flame thicknesses and the burning velocities are compared with that of free flames. The results show a non-monotonic dependence of the burning velocity on pressure considerably affected by the air/fuel equivalence ratio, which is well depicted by the models used in this work with some identified limitations. By performing a parametric study with the 1D model, dispersion of heat and mass has been identified as the major responsible mechanism of increasing of burning velocity with pressure. The DPLS results showed considerable temperature variations along the space coordinate, which were not considered up to now in 1D models when calculating reaction rates and which could be the reason of the observed limitations.

The research work of the present study is focused on the numerical simulation of primary breakup of high-viscosity non-Newtonian fluids. For the experimental investigation of fluid properties such as viscosity, surface tension and flow behaviour on the jet breakup an external mixing twin-fluid nozzle is used, as investigated in a previous study [Müller et al., ASME Turbo Expo 2016, GT2016-56371]. To describe the disintegration process of the fluids, characteristic features like liquid jet morphology, breakup length, breakup frequency and spray angle are evaluated. Furthermore, the primary breakup of slurries is simulated without discretizing the particles as a third phase, which heavily reduced the computational effort. Instead, the physical properties (density, viscosity) of the liquid phase take the influence of the particles into account. The primary breakup was investigated using the open source CFD software OpenFOAM. To gather the morphology of the primary breakup and the flow field characteristics compressible large eddy simulations (LES) were performed and the movement of the gas-liquid interface was captured by means of the Volume of Fluid-Method (VOF). The conducted simulations showed good agreement with experimental results with respect to the characteristic features (e.g. breakup length) and the significant influence of viscosity and surface tension on the primary breakup. It is reasonably justified that the used OpenFOAM code and VOF is sufficient to simulate the primary breakup of particle laden liquids without discretizing particles as a third phase. Moreover, those findings contribute to a better understanding of the physics responsible of the breakup of high-viscosity liquid jets and as well to create an experimentally validated CFD based tool for future burner development and optimization.

In direct numerical simulation of turbulent combustion, the majority of the total simulation time is often spent on evaluating chemical reaction rates from detailed reaction mechanisms. In this work, an optimization method is presented for speeding up the calculation of chemical reaction rates significantly, which has been implemented into the open-source CFD code OpenFOAM. A converter tool has been developed, which translates any input file containing chemical reaction mechanisms into C++ source code. The automatically generated code allows to restructure the reaction mechanisms for efficient computation and enables more compiler optimizations. Additional performance improvements are achieved by generating densely packed data and linear access patterns that can be vectorized in order to exploit the maximum performance on HPC systems. The generated source code compiles to an OpenFOAM library, which can directly be used in simulations through OpenFOAM’s runtime selection mechanism. The optimization concept has been applied to a realistic combustion case simulated on two peta-scale supercomputers, among them the fastest HPC cluster Hazel Hen (Cray XC40) in Germany. The optimized code leads to a decrease of total simulation time of up to 40% and this improvement increases with the complexity of the involved chemical reactions. Moreover, the optimized code yields good parallel performance on up to 28,800 CPU cores.

The research work of the present study is focused on the influence of design parameters of twin-fluid nozzles used for the atomization of high-viscosity fuels with respect to the primary breakup of the liquid jet. Two external mixing twin-fluid nozzles, which have already been investigated in previous studies [1, 2], were chosen as basic design. Based on the previous findings the web thickness between fuel and oxidizer supply was varied. In addition both designs were extended by a channel for internal mixing of gas and liquid with a length to diameter ratio of one. Moreover one of the basic nozzles was scaled by decrease of the effective areas in a way that momentum flux ratio as well as gas to liquid mass flow ratio was kept constant. The newly designed atomizers were subsequently investigated with regard to the influence of the changes upon the primary jet breakup using CFD simulations. The numerical simulations were conducted by means of the open source package OpenFOAM. The Volume of Fluid method was used for the determination of the gas-liquid interface. These simulations were then compared with experimentally validated simulations of the basic nozzle designs with regard to the breakup morphology of the jet and the mode of the primary surface instability. In addition, the liquid structure was examined by comparison of breakup length and frequency. The results of these simulations showed that small changes in the atomizer design heavily influence the primary breakup, which in turn influences the overall performance of the atomizer (e.g. SMD). Moreover, these findings will contribute to a better understanding of the physics of the breakup of high-viscosity liquid jets and as well to create an experimentally validated CFD based tool for future burner development and optimization.

Emissions characteristics of lean, turbulent, partially premixed swirled flames of synthetic fuels along with a standard Jet A-1 fuel are studied. The investigated synthetic fuels are (a) Fully synthetic jet fuel (FSJF), (b) Fischer Tropsch synthetic paraffinic kerosene (FT-SPK), (c) FT-SPK+20 % hexanol, and (d) FT-SPK+50 % naphthenic cut. The measurements are performed in a tubular combustor equipped with a burner based on the principle of air-blast atomization. The exhaust gas compositions are measured using a non-dispersive infrared gas analyzer for carbon dioxide (CO2) and carbon monoxide (CO), a flame ionization detector for unburned hydrocarbons (UHC), and a chemical luminescence detector for nitric oxides (NO and NO2). The emissions indices (EI) of CO and NOX of the investigated fuels are calculated using guidelines provided by the Society of Automotive Engineers (SAE). Measurements are performed at several combustor pressure levels, i.e., 0.3, 0.54 and 0.8 MPa, to compare the emissions behavior of the investigated fuels at varied operating conditions. At 0.3 MPa of combustor pressure, the order of fuels with their increasing formation of NOX are FSJF, FT-SPK+20 % hexanol, Jet A-1, FT-SPK+50 % naphthenic cut and neat FT-SPK. Differences in the observed NOX formation behavior of the investigated fuels are attributed to their probable different degrees of mixing with air in the combustor. At 0.8 MPa, no significant differences in their emissions characteristics are observed due to very low absolute values; hence we report that at higher pressure conditions which prevail in the aero-engine combustion systems, the emissions characteristics of tested synthetic fuels are very close to that of standard Jet A-1 fuel.

Auto-ignition and flame propagation are the two different controlling mechanisms for stabilizing the flame in secondary stage combustion in hot vitiated air environment and at elevated pressure. The present work aims at the investigation of the flame stabilization mechanism of flames developing in such an environment. In order to better understand the structure of turbulent flames at inlet temperature well above the auto-ignition temperature, the behavior of laminar flames at those conditions needs to be analyzed. As an alternative to challenging and expensive measurements at high temperature and pressure, the behavior of laminar flames at such conditions can be predicted from theory using mathematical simulation. In the present work, the laminar burning velocities and flame structures of premixed stoichiometric methane/air mixtures for inlet temperatures from 300 to 1450 K and absolute pressures from 1 to 8 bar have been calculated using a freely propagating laminar, one dimensional, planar flame model. The prediction shows that at inlet temperatures below the auto-ignition temperature, the predicted laminar burning velocity which corresponds to the unburned mixture velocity in order to create a steady laminar flame decreases with increase in pressure. When the inlet temperature of the mixture goes well beyond the auto-ignition temperature of the mixture, however, the unburned mixture velocity increases steeply at higher pressure level, because of a complete transition of the flame structure.

Turbulent flames are utilised in almost all applications employing combustion. Thus, better understanding of the combustion process is the key for increasing the efficiency of fuel consumption and reducing the environmental footprint of future combustors. As the turbulent flame velocity depends on the laminar burning velocity and intensity of the turbulence, the laminar burning velocity can be considered as a crucial parameter for fuel characterisation. Since turbulent flame can be considered basically as wrinkled laminar flame (flamelet approach), the effect of stretch (caused by turbulence) on the flame propagation must also be considered. The Markstein number, which quantifies the response of a laminar flame to the stretch, can be employed for this purpose. Additionally, the Markstein number can be utilised as an indicator for laminar and turbulent flame front stability. The experimental data of the conventional liquid fuel – kerosene Jet A-1 – are scarce, especially at elevated pressure conditions. Identification of the parameters for elevated pressures is very important as such conditions are closer to real operational ones. Additionally, as the laminar velocity embodies fundamental information on the diffusivity, reactivity and exothermicity of a given mixture, it is often utilized for the validation of fuel reaction mechanisms. The combustion characteristics laminar burning velocity and Markstein number of kerosene Jet A-1 are investigated experimentally in an explosion vessel. For this purpose an optical laser method is employed based on the Mie-scattering of the laser light by smoke particles. Within this experimental study the influence of three crucial parameters: initial temperature, initial pressure and mixture composition on the burning velocity and Markstein number are investigated. The experiments are performed at five different pressures: 0.1, 0.2, 0.4, 0.6 and 0.8 MPa; three different temperatures: 373, 423 and 473 K; and for a range of equivalence ratios of 0.67–1.43. Along with the experiments, two different reaction mechanisms are used to evaluate their ability to predict the experimentally observed laminar burning velocities. The observed results are compared and discussed in detail.

The jet in crossflow is a configuration of highest theoretical and practical importance, in which the turbulent mixing plays a major role. High-resolution measurements using Particle Image Velocimetry combined with Laser Induced Fluorescence have been conducted and used to validate simulations ranging from simple steady-state Reynolds-averaged Navier Stokes to sophisticated large-eddy simulation. The reasons for the erratic behavior of steady-state simulations in the given case, in which large-scale structures dominate the turbulent mixing, have been discussed. The analysis of intermittency proved to be an appropriate framework to account for the influence of these flow structures on the jet in crossflow, contributing to the explanation of the poor performance of the steady-state simulations.

A new method to analyze fluid flow in open-cell foams is given by the rise in computational speed, which makes it possible to calculate the heat and mass transport within these structures even with conventional CFD methods. In the present study, air flow through several structures has been explicitly calculated using the standard Navier–Stokes equations. Together with the flow, the dispersion of an initial tracer jump is tracked through the structures and analyzed with respect to effective diffusivity coefficients. The structures comprises a broad range of open-cell ceramic foams which were reconstructed from tomographic data obtained by means of MRI and μ - CT scans.

One of the key factors that affect the combustion performance in gas turbines and therefore the amount of pollutant emission and the efficiency is the atomization quality of the liquid fuel. Often reliable droplet size distributions for combustion simulations are missing. In this work a correlation for the Sauter mean diameter based on energy equilibrium is derived and compared to results from three-dimensional numerical simulation of the atomization of a liquid film and to an empirical correlation from the literature. It is shown that the derived correlation agrees better with the results of the simulations than the empirical correlation.

This study deals with the impact of the operating conditions, e.g. pressure, preheating temperature, pressure drop across the nozzle, nozzle size and stoichiometry, on the reaction zone location and spray evaporation progress in case of a lifted flame. Lifted flames are highly valued for their NOx reduction potential and for their low susceptibility to flash-back and thermo-acoustic instabilities. These advantageous features arise from the improved homogeneity of the fuel-air mixture provided to the reaction zone. One distinctive feature of the lifted flames is the presence of the so called lift-off zone located between nozzle outlet and main reaction zone. Within the lift-off zone fuel and oxidizer remain a certain time in contact and mix together prior to the onset of the combustion reaction. This leads to a more uniform heat release distribution and lowers the nitrogen oxides emissions at lean conditions by reducing the temperature spikes. In contrast to many other studies the subject of investigation was not a plain jet flame, but a modified version of the airblast nozzle, widely used in industrial applications. The nozzle was operated with liquid kerosene. As liquid fuels are easier to handle than gaseous or solid, it is expected that many efforts in the future will focus on the development of liquid fuels surrogates. Our previous investigations have shown, that the nozzle is well suited to be operated with gaseous fuels as well (Fokaides et al, J Eng Gas Turbine Power 130, 011508 2008). The position of the reaction zone was determined by means of chemiluminescence of the OH∗ radicals and from its location the lift-off height was derived. In addition the fuel evaporation progress was measured by means of light scattering, revealing that fuel droplets and main reaction zone are well separated. It was found that the operating conditions have a versatile impact on the length of the lift-off zone and spray cone and thus on the degree of pre-evaporation and premixing. Thus, it may be concluded, that through a proper choice of operating conditions and combustor size a desired lift-off height can be adjusted in accordance with criteria, like available space, required emission levels etc.