A priori analysis of combustion modeling closures for aviation engines using DNS of lab-scale combustor

Concern for emission reduction has motivated the development of new cost-effective alternative sustainable aviation fuels (SAFs). Drop-in SAFs with blends of the certified and the alternative fuels are beneficial because, ideally, they do not require engine modifications for use in current aviation engines. However, the employment of new jet fuels requires extensive testing and certification. Predictive and efficient combustion models are paramount to reduce costs related to the usage of SAFs in aviation engines. Flame stabilization and combustion dynamics are a product of competing processes related to the fuel chemical timescale, evaporation and mixing. Aero-engine combustion is often modeled with the flamelet approach based on mixture fraction, which was proven effective in predicting autoignition and stabilization of non-premixed jet flames. The complex flame dynamics encountered in aero-engines, such as multi-modal combustion, local extinction and reignition, challenge the current modeling closures. Multi-modal combustion can be captured with the correct closure for the dissipation terms in the flamelet equation based on mixture fraction and progress variable. On the other hand, extinction and reignition are also governed by the statistical nature of the process. The objective of this work is to perform an a priori analysis of the flamelet modeling closures using data obtained with a DNS of a lab-scale combustor. DNS simulations are performed in the low-Mach solver of the Pele Suite, PeleLMeX. Lagrangian multi-phase modeling is used to capture the liquid spray injection of Jet-A (reference fuel) and C1 as a representative of a low cetane number SAF. Adaptive Mesh Refinement (AMR) is used to enable a more efficient simulation of a more realistic domain size and embedded boundary treatment is used to model a bluff-body geometry.