Scalar Variance Model Validation for High-Speed Variable Composition Flows

Abstract

A scalar variance model for predicting local values of turbulent Prandtl and Schmidt number in high-speed flows is discussed. A series of fundamental validation problems is presented along with several illustrative applications in more complex flows. A model for turbulent Schmidt number is presented based on a mixture-fraction variance and dissipation rate approach. The model for turbulent Prandtl number is formulated from internal energy variance and its dissipation rate, and is presented in a companion paper. These scalar variance models are applied in parallel with a k-e turbulence model to provide local predictions for the length and time scales for turbulent scalar mixing. Model performance is evaluated by comparison of predictions to measured data for a series of fundamental nonreacting experiments. The validation cases demonstrate the models’ ability to capture the effects of non-uniform turbulent Prandtl and Schmidt number behavior in flows of increasing complexity. Application of the models to a mixing simulation with flow separation and three-dimensional effects is presented, showing model predictions in a geometry representative of a scramjet flow path/fuel injection configuration.