DLR Portal
Home|Sitemap|Contact|Accessibility Imprint and terms of use Privacy Cookies & Tracking |Deutsch
You are here: Home:Departments and Groups:Multiphase Flow and Alternative Fuels
Advanced Search
Key Research Topics
Departments and Groups
Chemical Kinetics and Analytics
Computer Simulation
Gas Turbines
Combustion Diagnostics
Multiphase Flow and Alternative Fuels
High-pressure Experiments
Research Facilities
Projects
Publications
Job Opportunities
Press Release
How to find us
Research Field
Print

Fuel placement in gas turbine combustion chambers



Gas turbines still play an important role in aviation and the combustion chamber is an essential part of this technology. Environmental sustainability and requirements such as efficiency, safety and reliability are important parameters defining the performance of today's gas turbines. An inefficient combustion not only leads to higher fuel consumption but also to significant pollutant emissions.

For this reason, the fuel placement in combustion chambers is of particular importance and thus subject of current research projects. In order to adequately model the combustion process, the individual sub-processes in the combustion chamber have to be carefully analyzed and understood. The most important sub-processes in a modern staged combustion system are illustrated in the figure. Generally, the same sub-processes can be identified in both stages. Liquid is introduced via an injector. During the atomization, the fuel disintegrates into small droplets. As a high volume fraction is occupied by the liquid phase, collisions, agglomeration and break-up phenomena are of importance. Subsequently, the resulting fuel droplets spread in the combustion chamber and interact with turbulent structures in the gas phase. During this turbulent dispersion, the droplets heat up and vaporize. Finally, the fuel vapor reacts with the oxygen in the air. The flame is stabilized by means of large-scale flow recirculations. These recirculations aim at generating a low-velocity region where the flame is able to anchor. Additionally, the recirculatory flow transports hot combustion products upstream to continuously ignite the fresh fuel-air mixture. These mixing processes determine the conversion rates and emissions.

Reference:

G.Eckel, Large Eddy Simulation of turbulent reacting multi-phase flows, VT-Forschungsbericht 2018-2

 

Atomization

During the atomization, the fuel disintegrates into small droplets. This fuel preparation sub-process pursues the objective of increasing the surface to volume ratio, which is beneficial for vaporization and thus the formation of an ignitable mixture. The MAT department conducts research on accurate and universal primary atomization models using Direct Numerical Simulation (DNS), which are applicable independent of the injector concept. However, these methods are still computationally intensive and very expensive. Therefore, experimental data or semi-empirical models are mostly used to generate the spray starting conditions in combustion simulations. In the department, such models were developed and validated for pressure-swirl atomizers and jet-in-crossflow atomizers. Airblast atomizers were also modeled. In contrast to purely empirical correlations, which generally only provide a global characteristic diameter, these semi-empirical models are able to deliver the sizes, velocities and starting positions of the droplets. Figure 1 shows a series of snapshots of the disintegrating liquid kerosene jet injected into the high-velocity cross-flow and further the application of the transient jet-in-crossflow atomization model in a generic premixing duct at ambient temperature, increased pressure (5.8 bar) and an air speed of 100 m / s. The experiment was performed at the DLR Institute of Propulsion Technology.

Figure 2 demonstrates the high accuracy of the model prediction with respect to the jet penetration for different boundary conditions. Usually, a combination of different atomizer types is applied in modern combustion chambers. For this reason, it is important to develop sub-models which are able to numerically reproduce the combined atomization process.

Reference:

G. Eckel, M. Rachner, P. Le Clercq, and M. Aigner, Semi-Empirical Model for the Unsteady Shear Breakup of Liquid Jets in Cross-flow, Atomization and Sprays, 2016, 26, 687-712

 

 

 

Dispersion

After the stage of atomization, the physical system consists of discrete liquid droplets with different sizes, which spread in the turbulent, continuous gaseous phase. The droplet trajectory depends on the relative velocity between the droplet and the surrounding gas and is influenced by the turbulent eddies. However, due to the fact that droplet sizes are generally smaller than the cell size of the computational grid in the CFD, the droplet movement cannot be directly calculated. For this reason, dispersion models are required for the droplet-turbulence interaction. Such models are developed and tested in the MAT department.

Figure 1 shows a LES simulation of the dispersion of glass particles in a swirling flow. The gas phase was calculated by the DLR in-house pressure-based solver THETA (link). The movement of the glass particles was simulated using the CFD code SPRAYSIM (link), which was developed at DLR and enables Lagrangian particle tracking on 3D unstructured grids.

Reference:

G. Eckel, M.  Rachner, P. Le Clercq, and M. Aigner, Assessment of two particle dispersion models for Large Eddy Simulations of confined swirling flows, ICMF 2019 - 10th International Conference on Multiphase Flow, 2019

 

 

Vaporization

In order to successfully ignite and accomplish a stable combustion, a sufficient amount of gaseous fuel must be present in the combustion chamber. The change of state from the liquid to the gaseous phase is called vaporization. The vaporization process has a significant influence on the ignition behavior, the stability limits, the thermal efficiency, and the pollutant emissions. Therefore, accurate modeling of the vaporization process is crucial to predict the performance of a combustor. Modeling vaporization means finding an accurate numerical description of the non-linear coupled processes of momentum exchange, heat and mass transfer. As aviation fuels are complex multi-component blends, the mixing in the droplet and the different volatility of the individual components have to be considered, too. In the case of real fuels such as Jet A-1 and alternative aviation fuels such as the Hydro-processed Esters and Fatty Acids (HEFA), efficient methods like the Continuous Thermodynamics Model (CTM) are used to map the hundreds of components of the fuel without losing relevant fuel composition information. Furthermore, the complex behavior of non-ideal fuel mixtures (e.g. blends of alcohols and hydrocarbons), as well as the behavior of heterogeneous mixtures such as emulsions, are accounted for by suitable models. Heterogeneous mixtures can cause micro-explosions due to the sudden vaporization of fuel components inside the droplet. Micro-explosions have a very high potential for heavy oils or highly viscous fuels because they lead to a secondary atomization of the drops.

To ensure the reliability of the developed models, the accuracy of the vaporization models are determined and documented on the basis of comparisons with experimental data for a large number of different fuels. Experimental studies were carried out, for example, in cooperation with the Instituto Motori of the Consiglio Nazionale delle Ricerche in Naples in a preheated, turbulent flow under atmospheric conditions. Figure 1 shows the predicted development of the mean diameter D10 and the standard deviation DRMS of the spray diameter distribution in comparison to experimental data of a kerosene spray.

Reference:

S. Ruoff, B. Rauch, P. Le Clercq, and M. Aigner,  Assessment of the Comparability of Droplet Evaporation Fuel Sensitivities between a Unit Test Case and an Aviation Gas Turbine Combustor. AIAA Scitech Forum, 2019


Research Fields
Fuel placement in gas turbine combustion chambers
Alternative Fuels
Development of methods for the simulation of multi-phase flows
2019
Software
SimFuel platform
SPRAYSIM
Related Topics
Propellants and Fuels
Dependability
Pollutant
Copyright © 2022 German Aerospace Center (DLR). All rights reserved.