报告题目:Multi-timescale and Correlated Dynamic Adaptive Chemistry and Transport (CO-DACT) Modeling of Ignition and Flame Propagation of Jet Fuel Surrogate Mixtures 

 

报告人: Prof. Yiguang Ju 

Department of Mechanical and Aerospace Engineering, Princeton University, Princeton, NJ -08544, United States 

*Corresponding Author Email: yju@princeton.edu 

 

时间：2015年8月20日(周四) 9:30 

 

地点：中科院力学所1号楼312会议室 

 

报告摘要： 

A correlated dynamic adaptive chemistry and transport (CO-DACT) method is developed based on our previous correlated dynamic adaptive chemistry (CO-DAC) method to further improve the computational efficiency of the transport coefficients such as the mass diffusivities, heat conductivities, and viscosities. The concept of the correlated groups in both time and space coordinates for chemistry and transport is proposed by using a few key phase parameters which dominate the chemistry pathways and transport coefficients. Correlated reduced chemistry and transport coefficients are updated dynamically by specifying different threshold values of phase parameters of correlated groups. For transport, the mixture averaged diffusion model is applied to calculate the transport coefficients based on the correlated groups. Only one calculation of the transport coefficients is required for all the computation cells in the same correlated group. The advantages of the CO-DACT method are that it not only provides the flexibility and accuracy for the calculation of chemistry and transport coefficients for a large kinetic mechanism but also avoids redundant calculations in time and space when the chemistry pathways and the transport coefficients are correlated due to the similarities in phase space. The simulations of premixed propagating spherical flames as well as one-dimensional diffusion flames of a jet surrogate fuel are carried out to validate the proposed algorithm. The impact of the selection of the phase parameters as well as the influence of the threshold value at various pressures and equivalent ratios will be examined in this paper. The results show that the present CO-DACT method is not only computationally efficient (faster by two-orders of magnitudes) but also robust and accurate for large kinetic mechanisms.