Large eddy simulation of turbulent partially premixed dimethyl ether jet flame by the extended direct moment closure model coupled with acceleration algorithms

Developing accurate and efficient turbulent combustion models for partially premixed flames remains a key challenge due to the coexistence of premixed and non-premixed flames. In the present study, the direct moment closure (DMC) model, which is methodologically applicable to all combustion regimes, has been extended to multi-reactant reactions in complex chemistry, along with the incorporation of various acceleration algorithms to reduce computational cost. A TNF Workshop benchmark flame, i.e., the piloted partially premixed DME jet flame is simulated by the extended model, and the predicted results are compared with the experimental data and those of the typical conditional moment closure (CMC) model and the transported probability density function (TPDF) model. It is found that the extended DMC model has the capability of accurately predicting the partially premixed DME jet flame and better performance than the CMC and TPDF models. Among the acceleration algorithms, the tabulated dynamic adaptive chemistry method achieves the best performance, with all RMS errors below 2.2 % and an acceleration ratio of 2.22. These results demonstrate that the DMC model coupled with the TDAC acceleration algorithm is promising for turbulent combustion, offering both high accuracy and computational efficiency.