TY - GEN A1 - Gerber, Stephan A1 - Oevermann, Michael T1 - A two dimensional Euler-Lagrangian model of wood gasification in a charcoal bed — Particle histories T2 - Powder Technology N2 - An Euler-Lagrangian simulation applied to wood gasification in a fluidized bed is used to investigate individual particle histories and to provide a statistical analysis for temperature, apparent density as well as radial and axial positions of charcoal and dry wood particles. The model and several parameter variations have been discussed in detail in previous articles [1], [2], [3]. Based on simulation results we find much higher particle heating rates than typically used to measure kinetic data for, e.g., pyrolysis models. Although we presented a rather complex interplay of particle heating rate, biomass decomposition, particle shrinkage and particle fluidization the simulation results emphasize the importance of the usage of realistic particle size distributions for the initial bed as well as the fuel inlet. Furthermore, particle shrinkage data in combination with mass decomposition data appear to be crucial for realistic simulations. KW - Wood gasification KW - Pyrolysis KW - Discrete element KW - Fluidized bed KW - Charcoal bed KW - Euler-Lagrange Y1 - 2018 U6 - https://doi.org/10.1016/j.powtec.2017.10.032 SN - 0032-5910 VL - 324 SP - 5 EP - 15 ER - TY - GEN A1 - Gerber, Stephan A1 - Oevermann, Michael T1 - A two dimensional Euler-Lagrangian model of wood gasification in a charcoal bed - Part III: parameter influence and comparison T2 - Powder Technology N2 - The efficient utilization of biomass in fluidized bed reactors depends on numerous operation conditions and parameters which can be investigated systematically with computational models. Based on a previous study [1] we compare a base scenario which mimics the experimental situation with different parameter settings for an Euler-Lagrangian simulation of wood gasification in a fluidized charcoal bed. For the varied parameters superficial velocity, reactor wall temperature, air inlet temperature, wood particle size, and the wood inlet temperature we analyse the simulation results based on data for temporal evolution of reactor outlet temperature, averaged particle temperature, overall wood mass, overall charcoal mass, concentrations of several gaseous species (N2, CO, CO2, H2, H2O, CH4, C2H2, and three virtual tar components) and axial barycenter data of particles bed mass. Furthermore we show time averaged data for gaseous species and gas phase temperature at the reactor outlet. At the end we critically examine our findings under consideration of the models opportunities and limitations. KW - Wood gasification KW - Fluidized bed KW - Euler-Lagrange KW - Discrete element method KW - Charcoal bed KW - Gasification Y1 - 2017 U6 - https://doi.org/10.1016/j.powtec.2016.12.057 SN - 0032-5910 VL - 310 SP - 163 EP - 174 ER - TY - GEN A1 - Wadekar, Sandip A1 - Janas, Peter A1 - Oevermann, Michael T1 - Large-eddy simulation study of combustion cyclic variation in a lean-burn spark ignition engine T2 - Applied Energy N2 - Multi-cycle large-eddy simulation (LES) was performed to investigate combustion cyclic variability (CCV) in a single cylinder spark ignition engine with a homogeneous lean ( = 1.25) isooctane-air mixture. The aim was to obtain physical insights into the early stage of combustion and its influence on CCV. Propagation of the flame was modeled by a transport equation for the filtered flame surface density within the LES framework. The ignition process was represented by the imposed stretch spark ignition model (ISSIM-LES). Ten consecutive cold flow LES cycles followed by two initialization cycles (12 cycles in total) were used to perform the reactive simulations concurrently. The simulation results were compared with experimental data. Although the number of computed cycles was fairly low, the LES was able to reproduce the cyclic variability observed in experiments both quantitatively and qualitatively. Firstly, validation of the simulation was done by comparing measured pressure traces. Secondly, correlations between the timing of the 10% fuel burnt mass fraction with early flame kernel growth and initial-to-turbulent transition period (in which there was an asymmetric flame kernel that persisted through the early development periods) were determined. Thirdly, calculated results of the flame propagation were analyzed at two cross-sections (in swirl and tumble planes) of the combustion chamber, which highlighted differences in instantaneous flame structures and propagation characteristics between the fastest and slowest cycles. Good overall agreement was obtained between the measurements and simulation data. The results revealed that the instantaneous velocity and fluctuation of flows around the spark vicinity affect growth of the early flame kernel and cause combustion cyclic variability. KW - Large-eddy simulation KW - Combustion cyclic variation KW - Flame surface density KW - Ignition modelling KW - Lean combustion Y1 - 2019 U6 - https://doi.org/10.1016/j.apenergy.2019.113812 SN - 0306-2619 VL - 255 ER - TY - GEN A1 - Gerber, Stephan A1 - Oevermann, Michael T1 - A two dimensional Euler-Lagrangian model of wood gasification in a charcoal bed - Part II: parameter influence and comparison T2 - Particuology N2 - A Euler–Lagrangian simulation was employed for a comprehensive parameter study of wood gasification in a fluidized charcoal bed. The parameters that were varied include the initial bed temperature, fuel mass flow rate, inert tar fraction, and kinetic energy losses caused by particle–particle and particle–wall collisions. The results of each parameter variation are compared with a base scenario, previously described in detail in Part I of this study (Gerber & Oevermann, 2014). The results are interpreted by comparing the reactor outlet temperature, averaged particle temperature, overall wood mass, overall charcoal mass, concentrations of several gaseous species, and axial barycenter data for particles obtained with different sets of parameters. The inert tar fraction and fuel mass flow rate are the most sensitive parameter, while the particle–particle and particle–wall contact parameters have only a small impact on the results. Increasing the reactive tar components by 19% almost doubled the amount of reactive tars at the reactor outlet, while decreasing the restitution coefficients of the particle collisions by 0.2 results in higher overall gas production but almost no change in bed height. Herein, our numerical results are discussed in detail while assessing the model restrictions. KW - Wood gasification simulation KW - Gas–solid flow KW - Discrete element method KW - Pyrolysis KW - Gasification KW - Charcoal Y1 - 2017 U6 - https://doi.org/10.1016/j.partic.2017.01.004 SN - 1674-2001 VL - 35 SP - 22 EP - 30 ER - TY - GEN A1 - Lackmann, Tim A1 - Hewson, John A1 - Knaus, Robert A1 - Kerstein, Alan R. A1 - Oevermann, Michael T1 - Stochastic modeling of unsteady extinction in turbulent non-premixed combustion T2 - Proceedings of the Combustion Institute N2 - Turbulent fluctuations of the scalar dissipation rate have a major impact on extinction in non-premixed combustion. Recently, an unsteady extinction criterion has been developed (Hewson, 2013) that predicts extinction dependent on the duration and the magnitude of dissipation rate fluctuations exceeding a critical quenching value; this quantity is referred to as the dissipation impulse. The magnitude of the dissipation impulse corresponding to unsteady extinction is related to the difficulty with which a flamelet is exintguished, based on the steady-state S-curve. In this paper we evaluate this new extinction criterion for more realistic dissipation rates by evolving a stochastic Ornstein–Uhlenbeck process for the dissipation rate. A comparison between unsteady flamelet evolution using this dissipation rate and the extinction criterion exhibit good agreement. The rate of predicted extinction is examined over a range of Damköhler and Reynolds numbers and over a range of the extinction difficulty. The results suggest that the rate of extinction is proportional to the average dissipation rate and the area under the dissipation rate probability density function exceeding the steady-state quenching value. It is also inversely related to the actual probability that this steady-state quenching dissipation rate is observed and the difficulty of extinction associated with the distance between the upper and middle branches of the S-curve. KW - Extinction KW - Unsteady flames KW - Non-premixed flame KW - Scalar dissipation rate KW - Turbulence Y1 - 2016 U6 - https://doi.org/10.1016/j.proci.2016.07.014 SN - 0082-0784 VL - 36 IS - 2 SP - 1677 EP - 1684 ER - TY - GEN A1 - Somhorst, Joop A1 - Uczak De Goes, Wellington A1 - Oevermann, Michael A1 - Bovo, Mirko T1 - Experimental Evaluation of Novel Thermal Barrier Coatings in a Single Cylinder Light Duty Diesel Engine T2 - SAE Technical Papers N2 - The objective of this investigation was to improve the thermal properties of plasma sprayed thermal barrier coatings (TBC) for internal combustion engines. There is a need for further reduction of thermal conductivity and volumetric heat capacity and the negative effects on heat loss and combustion phasing of surface roughness and permeable porosity, typical for plasma sprayed coatings, should be minimized. Four measures for improvement of TBC properties were evaluated: i) modification of the coating's microstructure by using a novel suspension plasma spraying method, ii) application of gadolinium-zirconate, a novel ceramic material with low thermal conductivity, iii) polishing of the coating to achieve low surface roughness, and iv) sealing of the porous coating surface with a polysilazane. Six coating variants with different combinations of the selected measures were applied on the piston crown and evaluated in a single cylinder light duty diesel engine. The experimental data was modeled with multiple linear regression to obtain confidence intervals for the measurement results and to correct the data for variations of surface roughness, combustion phasing and compression ratio for the different pistons. The main tool for evaluation of the coating properties was cylinder pressure analysis, providing the apparent rate of heat release, indicated efficiency, wall heat loss, and exhaust loss. The new TBC microstructure from suspension plasma spraying in combination with the use of gadolinium-zirconate showed promising results with respect to indicated efficiency and heat loss reduction. Y1 - 2019 U6 - https://doi.org/10.4271/2019-24-0062 SN - 2688-3627 SN - 0148-7191 ER - TY - GEN A1 - Wadekar, Sandip A1 - Oevermann, Michael A1 - Lipatnikov, Andrei T1 - Large Eddy Simulation of Stratified Combustion in Spray-guided Direct Injection Spark-ignition Engine T2 - SAE Technical Papers N2 - Stratified combustion in gasoline engines constitutes a promising means of achieving higher thermal efficiency for low to medium engine loads than that achieved with combustion under standard homogeneous conditions. However, creating a charge that leads to a stable efficient low-emission stratified combustion process remains challenging. Combustion through a stratified charge depends strongly on the dynamics of the turbulent fuel-air mixing process and the flame propagation. Predictive simulation tools are required to elucidate this complex mixing and combustion process under stratified conditions. For the simulation of mixing processes, combustion models based on large-eddy turbulence modeling have typically outperformed the standard Reynolds averaged Navier-Stokes methods. Therefore, we investigated spray-guided stratified combustion in a single cylinder engine using large-eddy turbulence modeling with a variant of the flame speed closure (FSC) model for premixed turbulent combustion. This model reveals the influence of the mixture composition on the flame speed. The effect of fluctuations in the composition were accounted for by using a presumed probability density function (PDF) approach for the mixture fraction. The fuel injection process was modeled with a standard Lagrangian spray model. More importantly, the measured in-cylinder pressure traces for three different loading cases with varying injection and ignition timings (leading to different levels of stratification) were accurately reproduced by the simulation. High-speed video images were used to evaluate the ability of the model to accurately simulate flame propagation under stratified conditions. The influence of mixture fluctuations on flame propagation was also investigated. KW - Single cylinder engines KW - Ignition timing KW - Combustion and combustion processes KW - Fuel injection Y1 - 2018 U6 - https://doi.org/10.4271/2018-01-1420 SN - 2688-3627 SN - 0148-7191 ER - TY - GEN A1 - Fistler, Marco A1 - Kerstein, Alan R. A1 - Wunsch, Scott A1 - Oevermann, Michael T1 - Turbulence modulation in particle-laden stationary homogeneous shear turbulence using one-dimensional turbulence T2 - Physical Review Fluids N2 - Turbulence modulation in particle-laden stationary homogeneous shear turbulence (HST) is investigated using one-dimensional turbulence (ODT), a low-dimensional stochastic flow simulation model. For this purpose, an ODT formulation previously used to study turbulence modulation in forced homogeneous isotropic turbulence (HIT) is extended, so that the model emulates the anisotropic character of HST and, potentially, anisotropic turbulence in general. This is done by limiting the kinetic-energy redistribution during an eddy event to an exchange involving two velocity components, where the three possible choices of the omitted component define three eddy types whose relative likelihoods control the anisotropy. Comparisons of ODT and direct-numerical-simulation results with reference to signatures of turbulence modulation are the basis of a broader ODT parameter study that is reported. Owing to the reduced dimensionality of ODT, it is found that the fidelity of the model for single-phase HST does not extend to particle effects on flow anisotropy, but for quantities averaged over components, parametric trends are captured. The consistent approach to case comparisons that was introduced in the HIT study to evaluate sensitivities to particle-phase parameters in a given flow configuration is extended here to a cross-comparison of HST and HIT model results, and its efficacy is again confirmed. The results provide an overall characterization of the potential for ODT to support the incorporation of particle-induced turbulence modulation into subgrid-scale closures of large-eddy simulations. Y1 - 2020 U6 - https://doi.org/10.1103/PhysRevFluids.5.124303 SN - 2469-990X VL - 5 IS - 12 ER - TY - GEN A1 - Wadekar, Sandip A1 - Yamaguchi, Akichika A1 - Oevermann, Michael T1 - Large-Eddy Simulation on the Effects of Fuel Injection Pressure on the Gasoline Spray Characteristics T2 - SAE Technical Papers N2 - Increasing the injection pressure in gasoline direct injection engines has a substantial potential to reduce emissions while maintaining a high efficiency in spark ignition engines. Present gasoline injectors are operating in the range of 20 MPa to 25 MPa. Now there is an interest in higher fuel injection pressures, for instance, around 40 MPa, 60 MPa and even higher pressures, because of its potential for further emission reduction and fuel efficiency improvements. In order to fully utilize the high-pressure fuel injection technology, a fundamental understanding of gasoline spray characteristics is vital to gain insight into spray behavior under such high injection pressures. The understanding achieved may also be beneficial to improve further model development and facilitate the integration of such advanced injection systems into future gasoline engines. In the present study, a gasoline fuel spray has been investigated over a range of fuel injection pressures from 40 to 150 MPa through a numerical simulation study. The numerical calculations have been performed in a constant volume chamber under non-vaporizing conditions to best match the experimental setup. The numerical model utilized a large-eddy simulation (LES) approach for the gas flow and a standard Lagrangian spray model for the liquid phase. The spray atomization has been modeled using the Kelvin Helmholtz - Rayleigh Taylor (KH-RT) atomization model with a droplet size distribution from the injector assumed to follow a Rosin-Rammler distribution function. Simulation results for the spray liquid penetration length are validated with experimental findings under different fuel injection pressures. Afterwards, an arithmetic mean droplet diameter (D10) and a Sauter mean droplet diameter (D32) as a function of pressure are compared against the measured droplet diameters. Simulated drop size distributions are presented and compared with measured droplet sizes. The results indicate that a high fuel injection pressure increases the liquid penetration length and significantly reduces droplet sizes. The results also exhibit that the SMD decreases from 13.4 μm to 7.5 μm, when injection pressure changes from 40 MPa to 150 MPa and that probability of finding the 5-9 μm droplet diameter decreases from 72% to 40% for the injection pressure drops from 150 MPa to 40 MPa. KW - Spark ignition engines KW - Fuel injection KW - Fuel economy KW - Energy conservation Y1 - 2019 U6 - https://doi.org/10.4271/2019-01-0060 SN - 2688-3627 SN - 0148-7191 ER - TY - GEN A1 - Somhorst, Joop A1 - Oevermann, Michael A1 - Bovo, Mirko A1 - Denbratt, Ingemar T1 - A Method to Evaluate the Compression Ratio in IC Engines with Porous Thermal Barrier Coatings T2 - SAE Technical Papers N2 - The compression ratio is an important engine design parameter. It determines to a large extend engine properties like the achievable efficiency, the heat losses from the combustion chamber and the exhaust losses. The same properties are affected by insulation of the combustion chamber. It is therefore especially important to know the compression ratio when doing experiments with thermal barrier coatings (TBC). In case of porous TBCs, the standard methods to measure the compression ratio can give wrong results. When measuring the compression ratio by volume, using a liquid, it is uncertain if the liquid fills the total porous volume of the coating. And for a thermodynamic compression ratio estimation, a model for the heat losses is needed, which is not available when doing experiments with insulation. The subject of this paper is the evaluation of an alternative method to assess the compression ratio. It is based on motored cylinder pressure data like other thermodynamic methods but does not need a model for the heat losses. The validation and application of the method is done with data from experiments involving two types of porous TBCs, performed on a light duty single cylinder diesel engine. The results indicate that the proposed method accurately predicts the compression ratio for porous thermal barrier coatings. KW - Combustion chambers KW - Combustion and combustion processes KW - Coatings, colorants, and finishes KW - Insulation Y1 - 2018 U6 - https://doi.org/10.4271/2018-01-1778 SN - 2688-3627 SN - 0148-7191 ER - TY - GEN A1 - Wadekar, Sandip A1 - Oevermann, Michael T1 - Large-Eddy Simulation Study of Ultra-High Fuel Injection Pressure on Gasoline Sprays T2 - Flow, Turbulence and Combustion N2 - The development of gasoline spray at ultra-high injection pressures was analyzed using Large-Eddy simulation (LES). Two different nozzle hole geometries, divergent and convergent shape, were considered to inject the fuel at injection pressures ranging from 200 to 1500 bar inside a constant volume spray chamber maintained at atmospheric conditions. The discrete droplet phase was treated using a Lagrangian formulation together with the standard spray sub-models. The numerical results were calibrated by reproducing experimentally observed liquid penetration length and efforts were made to understand the influence of ultra-high injection pressures on the spray development. The calibrated model was then used to investigate the impact of ultra-high injection pressures on mean droplet size and droplet size distribution. In addition, the spray-induced large-scale eddies and entrainment rate were evaluated at different ultra-high injection pressures. Overall, simulation results showed a good agreement with available measurement data. At ultra-high injection pressures mean droplet sizes were significantly reduced and comprised very high velocities. Integral length scales of spray-induced turbulence and air entrainment rate into the spray were larger at higher injection pressure compared to lower ones. KW - Sprayinduced turbulence KW - Ultra-high injection pressure KW - Large-Eddy simulation KW - Air-entrainment KW - Gasoline spray Y1 - 2020 U6 - https://doi.org/10.1007/s10494-020-00231-0 SN - 1573-1987 SN - 1386-6184 VL - 107 (2021) IS - 1 SP - 149 EP - 174 ER - TY - GEN A1 - Pütz, Michele A1 - Movaghar, Amirreza A1 - Oevermann, Michael T1 - Numerical simulation of a gasoline spray using one-dimensional turbulence for primary atomization T2 - ICLASS 2018 - 14th International Conference on Liquid Atomization and Spray Systems N2 - Predictive and reliable simulations have the potential to constitute a valuable tool for the optimization of spray systems if accurate submodels are developed for the entire range of the governing processes. The primary breakup of the turbulent liquid jet is one the most important mechanisms in sprays, yet the least developed in terms of numerical modeling. The most accurate method to simulate primary breakup is the proper resolution of liquid-gas interfaces and turbulent flow structures. However, a wide range of relevant length and time scales implicate grid requirements that are often prohibitive for real engineering applications. The most widely used method in practice is still the representation of both the continuous liquid core and the dispersed phase by means of discrete Lagrangian particles evolving in and interacting with the Eulerian gas phase. The available models for primary breakup are mainly phenomenological and involve a number of empirical constants. The one-dimensional turbulence (ODT) model is an alternative stochastic approach to model turbulence in flows with a dominant direction of property gradients. The stochastic representation of turbulent eddies on a one-dimensional domain enables high resolution at moderate computational costs. Applications of ODT to atomization revealed a great potential in recent studies. The objective of the present study is to combine ODT as a primary breakup model with a conventional Eulerian-Lagrangian method for the further spray evolution in order to asses ODT as a submodel in full spray models. Our numerical investigations were conducted on the ECN spray G, a gasoline-like, evaporating spray. The results in terms of spray penetration are encouraging, though the applicability of ODT to the transient injection phase and effects on additional spray characteristics require further investigation. KW - Breakup KW - Numericsl simulation KW - ECN Spray G KW - ODT KW - One-dimensional turbulence modeling KW - Gasoline Y1 - 2018 UR - https://research.chalmers.se/publication/519110 ER -