Research Topic

Advanced Modelling of Flame-Wall Interaction and Heat Transfer in Combustion Systems

Introduction and Motivation

The modern combustion research faces two major objectives, the optimization of the combustion efficiency and the reduction of pollutants according to the new emission standards. The current tendency is towards using lean and stratified burning conditions, where lower emissions and improved performance can be achieved. However, the gain related to decreasing of the harmful pollutant formation might be invalidated by increasing the combustion instabilities, as the reliability of such systems can be a challenging issue. Besides different combustion regimes, presence of cold walls in combustion systems can have enormous influence on the flame stabilization and excess emissions of unburned hydrocarbons, where these effects are even more significant in today's compact combustion chambers and novel IC-engines. Experimental investigations as well as theoretical analysis have provided only limited amount of information about the behavior and the structure of the turbulent flame in such systems. Accordingly, gaining deeper understanding into the physics that controls emissions related to flame-wall interaction is essential to explore new solutions, reduce costs, and improve development efficiency. This can be achieved by means of numerical modeling. Although substantial advances have been accomplished in various areas, such combustion and turbulence modeling, numerical methods, parallel-computing, there are numerous additional requirements to be met in order to provide a reliable numerical design tool. This project aims at developing an advanced CFD tool that is able to compute different kind of combustion regimes including flame-wall interaction and heat transfer occurring in combustion systems, especially in lean-burn and stratified devices.

Illustration of the flame stabilization within the Darmstadt stratified burner. The instantaneous temperature field is shown together with the chemical source term of the carbon dioxide (black lines).

Method and Theory

The present work focuses on lean stratified combustion systems, which are commonly found in practical combustion devices, as well as on perfectly premixed and non-premixed conditions. Large Eddy Simulation (LES) technique is used for capturing of the unsteady processes connected to practical turbulent combustion. This approach represents the best compromise between practicality and physical accuracy, since the turbulent motion associated with the large, energy-containing eddies is computed directly, whereas the effects of the smallest ones are modeled.

Implementing the combustion chemistry into LES involves finding suitable reaction mechanism and solving filtered equation for each individual species in the reaction mechanism. It's proper capturing may require hundreds of species and thousands of reactions to be accounted for, which can lead to numerical difficulties. These complex chemical mechanisms cannot be directly described in a three-dimensional numerical simulation of turbulent combustion, keeping in mind an enormous requirement on computer resources. Therefore, within this work Flamelet-Generated Manifold (FGM) is used as a reduction strategy which aims at describing detailed chemistry by only a few controlling variables using tabulation. Even though using tabulated chemistry, the flame front is still unresolvable on the LES computational mesh. Therefore, in order to determine its position and structure, in addition to FGM, different combustion models have to be used. One such model is the well-established Artificially Thickened Flame (ATF), where the flame front is thickened and therewith resolvable on coarser grids. This model however is mainly developed for the premixed combustion regime and in order to compute different combustion regimes, corresponding models need to be employed. Therefore, development and implementation of the Monte Carlo Stochastic Field Method and combining it with the FGM reduced chemistry, wherein no particular flame structure is assumed and which allows application in all combustion regimes, is part of this work.

Key Research Area

Multi-Physics, Large Eddy Simulation, Combustion Modeling, Stratified Combustion

Contact

Amer Avdić
Dr.-Ing.

Address:

Wöhlerweg 2

D-64293 Darmstadt

Germany

Phone:

+49 6151 16 - 24401 or 24402

Fax:

+49 6151 16 - 24404

Office:

S4|10-210

Email:

avdic (at) gsc.tu...

 Print |  Impressum |  Sitemap |  Search |  Contact |  Privacy Policy
zum Seitenanfangzum Seitenanfang