14 Externe wissenschaftliche Einrichtungen
Permanent URI for this collectionhttps://elib.uni-stuttgart.de/handle/11682/15
Browse
Item Open Access Boundary layer response to combustion instabilities and associated heat transfer(2010) Panara, Daniele; Aigner, Manfred (Prof. Dr.-Ing.)The development of efficient and environmental sustainable combustion systems is critical in the nowadays economy. The efficiency of an energy cycle is proportional to the highest cycle temperature but unfortunately, due to the major mechanism of nitric oxide formation, there is a temperature trade off between improved cycle efficiency, material constraints and low emissions. This considerations have pushed combustion chamber technology towards lean premixed flames where the tendency is to reduce local temperature peaks making use of a well studied air management. Unfortunately this kind of innovative systems have shown to be prone to combustion instabilities and higher wall heat load. In the present work, making use of numerical simulations, the wall boundary layer response to combustion instabilities has been studied in order to asses the effect of flow pulsations on heat transfer. In Chapter 2 the status of the art of combustor simulation has been presented. A full scale burner has been simulated with a commercial CFD (Computational Fluid Dynamic) code. The results have shown that in such complex simulations and in the presence of combustion instabilities, the correct predictions of wall heat load rely not only on the correct modeling of air-fuel turbulent mixing, chemical reactions and heat radiation, but also depend strongly on the near wall turbulence treatment and on the correct solution of the conjugate solid-fluid heat transfer problem at the wall (which details are discussed in Chapter 3). The main interest of the work is however the study of the near wall turbulence and the associated heat transfer in the presence of flow unsteadiness. In Chapter 4 the fundamental equations for the solution of the thermo-Fluid dynamic problem in turbulent unsteady flows have been introduced. Moreover, in Chapter 5 some unsteady analytical solutions in simple channel and pipe configurations as well as turbulent channel and pipe flow heat transfer and viscous loss correlations have been presented. In Chapter 6 and 7, making use of simplified but well defined academic test cases, the accuracy of different turbulence models for the prediction of the wall heat transfer response in presence of thermo-acoustic instabilities has been discussed. The results have shown the clear limitation of the use of wall functions both in URANS and LES applications and discrepancies with some experimental results. The most interesting results are however presented In Chapter 8 where the turbulent near wall structures and the associated heat transfer were in detail investigated by means of pulsating channel flow DNS (Direct Numerical Simulation) simulations. The response of the turbulent kinetic energy (k) and fluctuating temperature variance (k_theta) as well as their dissipation rates (epsilon and epsilon_theta) were reported at different flow pulsations and amplitudes. In order to highlight the most critical (for the turbulence models) flow conditions, the pulsation amplitudes and frequencies were chosen so to span different pulsating flow regimes. The DNS results have shown very complex turbulence unbalance phenomena depending on the flow pulsation and amplitude. Each term on the budget equation of k, k_theta and epsilon_theta have shown different answer to pulsation frequency and amplitude. The mean and phase locked averaged turbulent Prandtl number seems to be affected by flow pulsation as well as the unsteady heat transfer. The pulsation frequency increase determines an unsteady heat transfer amplitude decrease. The flow pulsation amplitude seems to affect the overall mean heat transfer value increasing up to two times its steady value with the increasing of pulsation amplitude. The present DNS database represents a highly valuable and unique work which can serve as a reference for the development of innovative unsteady turbulent heat transfer models. Finally, in Chapter 9, starting from the DNS data, a new turbulent heat transfer closure has been proposed. The new model was capable, in all the flow condition studied, to correctly predict the unsteady heat transfer mean value and pulsating amplitude showing the added value of the present DNS database for the understanding of the near wall turbulence behavior and the associated heat transfer in presence of flow unsteadiness.Item Open Access Numerical modeling of ignition processes in single- and multiphase flows(2014) Boyde, Jan Michael; Aigner, Manfred (Prof. Dr.-Ing. )In this work a combustion model is developed and presented which is applicable to a wide range of conditions through modifications to the Turbulent Flame Speed Closure model. The model predictions show that for all examined test cases including single- and multiphase conditions, a satisfying agreement with available experimental data is achieved. This underlines the usefulness of numerical tools for the investigation of ignition processes in the context of aircraft engines.Item Open Access Numerical simulations of soot formation in turbulent flows(2008) Di Domenico, Massimiliano; Gerlinger, Peter (PD Dr.-Ing.)This work deals with the simulation of soot formation phenomena under gasturbine-like conditions. Main goal is the development of a reliable CFD simulation tool able to predict trends of soot formation under different operating conditions. A detailed, finite-rate chemistry combustion model is presented and validated. Since soot particles are the result of thousands of reactions involving hundreds of species, an efficient, new sectional approach for soot precursors and related reactions is chosen in this work. Effects of turbulent fluctuations in temperature and species concentration on the chemical reaction rate are included by employing an assumed Probability Density Function approach. Finally, the simulation of a semi-technical scale burner under gasturbine-like conditions will demonstrate the validity of the developed tools. Although discrepancies with experimental soot particle distributions are observed, numerical simulations are able to reproduce the pressure dependence of the peak soot volume fraction quite well.