Development of an integral LES model for turbulent premixed combustion at elevated pressures

点击翻译按钮获取中文摘要

Primary energy is still often supplied by combustion processes. In many technical applications, such as gas turbine combustion chambers or supercharged piston engines, combustion takes place at elevated pressure. Due to the high energy density of fossil fuels, they are without alternative for commercial aircrafts in the foreseeable future and at least medium-term, despite increasing electrification, piston engines will still be used in conventional or hybrid drives. In order to optimize the flow and combustion process as well as to reduce development times and to save costs, numerical flow simulation is becoming increasingly important. Specifically, the large eddy simulation (LES) approach has become an important tool for flow simulation in recent years and is already used for first industrial configurations. Instead of modelling all turbulent scales, as often done in the past, many physical processes can be resolved with LES. However, in existing LES subgrid combustion models the effect of pressure has rarely been taken into account, the models are often insufficiently validated regarding high-pressure combustion. The occurrence of hydrodynamic instabilities, which can lead to a stronger flame wrinkling, is also often ignored. Finally, it is becoming more and more apparent that different model terms to be closed interact strongly with one another as well as with the numerical scheme. These individual parts should not be considered independent of each other and care should be taken when they are replaced. The overall goal of this project is the development and validation of an integral LES model for high pressure turbulent premixed combustion. The project is divided into two closely interlinked parts. First, a DNS data base is created for turbulent planar and Bunsen flames at different pressures. Subsequently, the data is spatially filtered to produce pseudo-LES fields (a-priori analysis). In this way modeling approaches can be compared with unclosed terms and promising models can be identified. However, the assessment of a model can ultimately only take place in a real LES, the so-called a posteriori analysis. This is the scope of the second part of the project. The previously selected closure approaches are implemented in an LES flow solver and validated using DNS data and experimental data from literature. The LES will also show the interaction of different submodels. In a second application period, the analyses and the model development will be extended to situations with spatially varying air-fuel ratio. This will be done by building a DNS data base of stratified turbulent flames using detailed chemical mechanisms.