@phdthesis{,
    author = {Breda, Paola},
    title = {Reduced chemistry models for the numerical investigation of flow and heat transfer in methane combustion devices},
    editor = {},
    booktitle = {},
    series = {},
    journal = {},
    address = {},
    publisher = {},
    edition = {},
    year = {2021},
    isbn = {},
    volume = {},
    number =  {},
    pages = {},
    url = {},
    doi = {},
    keywords = {Large-eddy simulations, Reduced chemistry models, Reacting flows, Heat transfer, Methane combustion},
    abstract = {Combustion technologies based on hydrocarbons will still play a relevant role in the long-term scenario, especially in the transportation sector. Numerical tools for the computation of turbulent reacting flows are of significant importance in the development phase of such technologies. Nevertheless, simulations of hydrocarbon combustion are not without diffculties. By increasing the dimensions of the chemical reaction differential equations, the stiffness of the chemical system increases, posing a limit to the available computational resources. An additional challenge is given by wall-confined turbulent flows at high Reynolds numbers, requiring wall treatment. Particular care is needed for the near-wall modelling if the prediction of thermal loads at the combustor wall is of interest, as key parameter to quantify thermomechanical limits. The main contribution of this work is the investigation of suitable combustion models for LES applications of methane flames. Reduced chemistry mechanisms and tabulated chemistry databases are chosen to speed-up the computation. The turbulence-chemistry interaction (TCI) on the small scales is investigated for CH 4 /air partially premixed flame configurations. Two transported PDF approaches are implemented to deal with reaction-diffusion manifolds (REDIM), respectively the Eulerian Stochastic Fields (ESF) and the Multiple Mapping Conditioning (MMC). Wall heat flux predictions are investigated on a sub-scale  rocket combustion chamber operated with gaseous CH4/O2. The best trade-off is sought in order to correctly represent the heat transfer at wall, while keeping the computation cheap. Flamelet-based chemistry databases including heat losses are used to model the effects of Flame/Wall Interaction (FWI). Additional validation of the reduced chemistry models is performed on an experimental configuration featuring near-wall reactions of CH 4 in a crossflow of hot combustion gases, where an autoignition delay is observed. This work shows that the implemented models based on chemistry databases significantly reduce the requirements on computational power, providing a satisfactory accuracy in the results. Strong extinction/re-ignition effects can be well represented by the ESF-REDIM model, the table also showing potential for predicting autoignition delays. When using finite rate chemistry, the use of MMC is found to be advantageous compared to ESF, although satisfactory predictions are already obtained by neglecting the TCI model. Flamelet databases including enthalpy losses provide satisfactory wall heat flux predictions for a variation of chemical mechanisms and near-wall treatments, if the flame is not subjected to autoignition phenomena.},
    note = {},
    
    school = {Universität der Bundeswehr München},
    
}