@article{SONG2025, 
author = {Jiahui SONG and Long MIAO and Aiguo XU and Yanbiao GAN and Feng CHEN and Yugan LIAO and Xiao HOU},
title = {Rarefaction effect on non-equilibrium characteristics of laminar shock wave/boundary layer interaction},
year = {2025},
journal = {Chinese Journal of Aeronautics},
volume = {38},
number = {10},
keywords = {Rarefaction effects, Discrete Boltzmann method, Shock wave/boundary layer interaction, Thermodynamic non-equilibrium, Kinetic theory},
url = {https://www.sciopen.com/article/10.1016/j.cja.2025.103538},
doi = {10.1016/j.cja.2025.103538},
abstract = {A Discrete Boltzmann Method (DBM) with a Maxwell-type boundary condition is constructed to investigate the influence of rarefaction on laminar Shock Wave/Boundary Layer Interaction (SWBLI). Due to the complexity of compressible flow, a Knudsen number vector Kn, whose components include the local Knudsen numbers such as Knρ and KnU, is introduced to characterize the local structures, where Knρ and KnU are Knudsen numbers defined in terms of the density and velocity interfaces, respectively. Since first focusing on the steady state of SWBLI, the DBM considers up to the second-order Knρ (rarefaction/non-equilibrium) effects. The model is validated using Mach number 2 SWBLI and the necessity of using DBM with sufficient physical accuracy is confirmed by the shock collision problem. Key findings include the following: the leading-edge shock wave increases the local density Knudsen number Knρ and eventually leads to the failure of linear constitutive relations in the Navier-Stokes (N-S) model and surely also in the lower-order DBM; the non-equilibrium effect differences in regions behind the leading-edge shock wave are primarily correlated with Knρ, while in the separation region are primarily correlated with KnU; the non-equilibrium quantities D2 and D4,2, as well as the viscous entropy production rate ṠNOMF can be used to identify the separation zone. The findings clarify various effects and main mechanisms in different regions associated with SWBLI, which are concealed in N-S model.}
}