Public Defence Simon Gövert

Modelling the effects of heat loss and fuel/air mixing on turbulent combustion in gas turbine combustion systems

The increasing energy demand and the yet more restrictive current emissions regulations are forcing the development of more efficient thermal systems and combustion engines. As the main part of the global energy supply will still be based on fossil fuels during the next decades, but also because renewable energy of sustainable sources will be stored in liquid fuels, the understanding of the combustion process is essential, since the chemical energy of fuel is generally converted into thermal energy by combustion. Major challenges in the design of modern combustion engines include the reduction of pollutant emissions, increment of fuel flexibility, increasing cycle efficiency and flame stability. To achieve these goals, an accurate description of the interaction of turbulence, chemical reactions and thermodynamics is required. In this context, the use of advanced numerical simulations is becoming a fundamental tool to provide detailed insights into the physical processes at relatively low cost. However, due to the different time and length scales existing in the combustion process, taking into account detailed chemistry in numerical simulations is still a challenge and remains out of scope for many complex combustion applications due to limited computational resources.


The present study is concerned with the development and validation of a simulation framework for the accurate prediction of turbulent reacting flows at reduced computational costs. Therefore, a combustion model based on the tabulation of laminar premixed flamelets is employed, as in the cases of the Flamelet Generated Manifold (FGM) and Flame Prolongation of ILDM (FPI) methods. By compilation of several flamelets, the model is extended to account for the effects of heat loss and fuel/air mixing. While flamelets at different enthalpy levels are utilized to account for non-adiabatic chemical kinetics, premixed flamelets at varying equivalence ratios are combined to address non-premixed and partially premixed conditions. The tabulation is parametrized in terms of scalar controlling variables that are used to couple the chemistry with the fluid mechanics, namely the mixture fraction, reaction progress variable and a normalized enthalpy scalar. Closures are presented for Reynolds-Averaged Navier-Stokes (RANS) and Large Eddy Simulation (LES) based on presumed shape Probability Density Functions (PDF) to account for turbulence-chemistry interaction. The model is implemented in the High Performance Computing (HPC) multi-physics code Alya, which is based on the Finite Element Method (FEM) for spatial discretization. Transport equations are solved for the scalar variables that control the combustion chemistry along with a low-Mach number formulation of the Navier-Stokes equations.

The prediction capabilities of the proposed approach for perfectly premixed conditions are assessed based on the reacting flow field of a confined turbulent jet flame. The effect of different heat transfer mechanisms and thermal conditions for the chamber walls is investigated in detail using a Conjugate Heat Transfer (CHT) approach. The influence of radiation, convection and heat conduction over the solid walls is examined by comparing the gas temperature with reference experimental data. Finally, the effect of partial premixing and heat loss on the reacting flow field prediction is addressed based on a swirl stabilized gas turbine model combustor

In conclusion, the modelling approach presented in this thesis is specifically designed to meet the design challenges that are encountered in the development of improved combustion technologies. It reveals good prediction capabilities for premixed and partially premixed conditions with special consideration of heat loss and non-adiabatic chemical kinetics.