Mrema, Honest2021-05-172021-05-172021-03https://hdl.handle.net/11299/220135University of Minnesota Ph.D. dissertation. March 2021. Major: Aerospace Engineering. Advisor: Graham Candler. 1 computer file (PDF); vii, 86 pages.High fidelity methods to simulate high speed turbulent combustion are of interest to the advancement of hypersonic air-breathing propulsion systems.These methods need to be cost-effective and efficient in order to be of practical use. To this end, this work seeks to compare and contrast the performance of two combustion modeling approaches that may meet these criteria. The two turbulent combustion modeling approaches are the flamelet progress/variable (FPV) and the evolution-variable manifold (EVM). Both models use tracking variables to reduce the dimensionality of the species transport equations. The main tracking variables being mixture fraction, which tracks the mixing of the fuel and oxidizer, and progress variable, which tracks how far the combustion process has advanced. For high performance at affordable computational cost, the investigated models are utilized with a hybrid large eddy simulation/Reynolds-averaged Navier-Stokes approach along with low dissipation numerical schemes. As a test bed, both approaches are applied to simulate a reacting transverse jet in a supersonic crossflow experiment. we compare OH-PLIF signals obtained from the experiment and simulations. The comparative study revealed that both models exhibit similar burning regions. The flame structures from the EVM simulations resembled the experimental results more compared to the FPV model. For this version of the FPV model, a more sophisticated compressibility correction is needed before a comprehensive comparison can be accomplished. Significant effort was put into improving the compressibility correction for the progress variable source term. This effort yielded a novel scaling approach for the progress variable production rate. This new approach focuses on scaling the reaction progress rates to account for compressibility. Although this type of correction has several benefits, the method struggles with numerical stability. Other efforts to enhance the FPV model have shown that this particular model is sensitive to flamelet manifold boundary conditions and user inputs. This work has revealed that the FPV combustion model, for the application of high speed turbulent combustion, requires further study.encomputational fluid dynamicsevolution-variable manifoldflamelet progress/variablelarge eddy simulationsupersonic combustionturbulence-chemistry interactionsThesis or Dissertation