Zusammenfassung: | The work addresses the numerical simulation of pulverized coal combustion. The relevance results from the current as well as from forecasted future worldwide utilization of coal combustion for the satisfaction of energy demands. In the present work, a modeling approach was developed which is capable of accurately describing pulverized coal combustion processes. Insight into and an understanding of physical processes can be gained in a scientific context using this approach. For industrial purposes it can be utilized for an efficient design of coal combustion chambers.
Coal is a complex fuel. Multiple physical effects on a broad range of scales take place during its conversion which requires the application of diverse interacting models. In the present work, the relevant physical processes were described at first. Then, the modeling approach was explained. One of this work’s central aspects is the development and application of a chemistry tabulation method (FGM) for pulverized coal combustion. The table is based on premixed flamelets. Finite rate chemistry effects as well as non-adiabatic physics are considered. In terms of the latter, preheated fuel states were integrated. Heat transfer between the phases in the context of chemistry tabulation was developed. Furthermore, the model was extended by an additional mixture fraction, so that the mixing of oxidizer, volatiles, char conversion products and supplementary gaseous fuel is describable. As it is included in the definition of this second mixture fraction, the consumption of oxygen during char burnout is accounted for. The high temperature reaction of carbon monoxide is also captured. The data structure as well as the covered state space of the four-dimensional manifold had to be re-organized so that a fast access to the non-equidistantly tabulated variables in combination with low memory requirements can be ensured. This development was conducted in the context of the large eddy simulation (LES) technique. The developed model was further coupled with a turbulence-chemistry-interaction model. For the latter an artificially flame thickening (ATF) approach was chosen. In particular, the numerical treatment of the developed model was adapted to the modified time integration method of the ATF-model. An Eulerian-Lagrangian approach was employed for the description of the interaction between the gaseous and the dispersed phase. In this regard, emphasis was put on a correct implementation of the particle source terms that had to be formulated as convective fluxes which was due to the discretization scheme. A coupling between the Lagrangian phase and a solver for detailed chemical kinetics was also implemented. All implementations were made in the academic research code FASTEST. For the corresponding verification, generic test cases were simulated in which the consistency of the developed modeling approach with respect to mass, species and energy balances was verified. Furthermore, the combustion of single coal particles was simulated applying both, the developed model as well as detailed chemistry kinetics, whereby the latter served as a reference solution. Several modeling aspects could be validated this way. Overall, it could be demonstrated that the coal conversion process could be sufficiently described by the modeling approach.
The developed model was then validated in configurations of practical relevance. The numerically obtained conversion of single particles in an inert drop tube reactor was in good agreement with experimental results. The ignition as well as the chemical reaction of volatile matter was investigated in a premixed flat flame burner configuration. It was demonstrated that the numerical simulation reproduced essential aspects that were observed in the experiment. Also, a quantitative good agreement could be found with regard to ignition heights. The modeling approach was then validated by simulating a novel coal combustion chamber, in which the combustion of coal particles is assisted and stabilized by a swirled methane flame and complex physical effects interact with each other. With regard to the flow field and the chemical reaction, essential experimental findings related with physical mechanisms could be reproduced. An understanding of the particle reaction and its impact on the reacting flow field could be obtained in the simulations.
|
---|