DFG-Individual Grants Programmes

21 February 2018

Primary energy is still often generated by combustion processes. In many technical applications, such as gas turbine combustion chambers or supercharged piston engines, combustion takes place under increased pressure. Due to the high energy density of fossil fuels, there will be no alternative for the foreseeable future, e. g. to the passenger turbine for commercial aircraft, and at least in the medium term the piston engine will continue to be used in conventional or hybrid engines despite increasing electrification. Numerical flow simulation is becoming increasingly important for optimizing process control, reducing development times and cutting costs. In particular, Large Eddy Simulation (LES) has become an important tool for flow simulation in recent years and is already being used industrially in initial configurations. Instead of modeling across scales, many physical processes can be resolved here. However, pressure influence has seldom been taken into account in the development of existing combustion models or the models are insufficiently validated for high-pressure combustion. The occurrence of hydrodynamic instabilities during high-pressure combustion, which can lead to increased folding of flames, is also not taken into account. Finally, it is becoming increasingly clear that different model terms to be closed interact strongly with each other as well as with the numerical method and cannot be viewed and developed or even exchanged separately from each other. The global objective of the project is therefore to develop and validate an integral overall model for LES high-pressure premix combustion. The project is divided into two closely interlinked projects. First, a DNA database is created from turbulent planar flames and bunsen flames at different pressures. The data is then spatially filtered to generate pseudo-LES data (a-priori analysis). In this way, modeling approaches can be compared with open terms and promising models can be identified. However, the evaluation of a model can only be done in a real LES, the so-called a-posteriori analysis. This is where the second part of the project starts. The previously selected closure approaches are implemented in a LES flow solver and validated on the basis of DNA data and experimental literature data. In LES, the interaction of different submodels is particularly apparent.