Book Description

Turbulent flows involving the interaction of shock waves and density variations are present in diverse areas of science and engineering. Inertial confinement fusion (ICF), for example, is a theoretical technology which could lead to clean, inexpensive energy for thousands of years to come. The implications of ICF are enormous as are the engineering challenges associated with it. One primary obstacle is modeling and avoiding the hydrodynamic instabilities which grow on the surface of the ICF capsule and which can prevent fusion from occurring. The commercialization of space exploration and privately owned satellites has led to an unprecedented number of orbital launches, which is expected to double each year for the foreseeable future. Modern rocket propulsion systems must become more reliable and efficient to sustain this growth. In the design of the supersonic rocket nozzle, a primary consideration is the avoidance of large flow separation caused by shock waves and the resulting lateral forces or side loads generated. These side loads decrease the lifetime of the nozzle and destabilize the rocket. Due to the complexity of this physical process and the extreme conditions in which it occurs, limited experimental data are available and still much is not understood. Being able to accurately model turbulence and its interaction with shock waves and multi-species mixing has broad impact on these and other problems. In this thesis, numerical simulation is used to model these turbulent interactions in simplified configurations of the two aforementioned applications, ICF and rocket nozzles. Large-eddy Simulation (LES) is used to explore the multi-material mixing processes of the Rayleigh-Taylor instability and the Kelvin-Helmholtz instability. Analysis of the data reveal interesting physics which were not previously documented and which have implications to astrophysical phenomena (type 1a supernovae) and advanced energy technology (ICF). LES is used to simulate an over-expanded planar nozzle which contains a shock wave turbulent boundary layer interaction. A new numerical method for simulating this type of flow is described and tested on a variety of configurations to demonstrate its improved effectiveness and accuracy over previous methods. The LES approach is then used to study the source of shock wave instability in the nozzle. A mechanism is identified and verified by additional simulations and a parametric study of the nozzle configuration. A simple, reduced order model is proposed which captures the shock wave dynamics using a fraction of the computational cost of LES.