Scalar mass conservation in turbulent mixture fraction based combustion models through consistent local flow parameters

Abstract

Mixture fraction-based models are widely employed for predicting turbulent non-premixed combustion processes due to their cost-effectiveness and well-established subfilter closure. In these models, the transport of reactive scalars in physical space is decomposed into two components: scalar transport relative to mixture fraction and transport of mixture fraction in physical space. Conventional flamelet models do not consider that these two processes have to be formulated consistently, which can lead to scalar mass conservation errors. In the context of multiphase flows, scalar transport in mixture fraction space is governed by three conditional flow-dependent parameters: the conditional scalar dissipation rate, the conditional scalar diffusion rate, and the conditional spray source term. The evolution of mixture fraction in physical space is typically modeled using the presumed Filtered Density Function (FDF) approach. This paper introduces a novel formulation for the conditional flow parameters that aligns with the presumed FDF approach, thereby ensuring scalar mass conservation. The proposed model is applied to a Large-Eddy Simulation (LES) of the inert ECN Spray A case, with a comparison against a conventional flow parameter model that employs an inverse error function shape for the scalar dissipation rate. The results indicate that the conventional model produces similar conditional dissipation rates to the new model in regions where combustion takes place. However, significant discrepancies are observed in the conditional diffusion rate, highlighting the susceptibility of the conventional model to scalar mass conservation errors for non-unity Lewis number scalars.