arXiv · 2611.20302
Population structure of the velocity distribution in normal shock waves: Gaussian representations, Mach-number similarity, and prediction
Abstract
Normal shock waves are commonly described as mixtures of upstream and downstream Maxwellians. Discrete-velocity solutions of three relaxation models up to Mach 12 and direct simulation Monte Carlo (DSMC) of hard-sphere and variable-hard-sphere gases up to Mach 25 are used to examine the population structure, Mach-number similarity and upstream precursor of their velocity distributions. Three anisotropic Gaussian populations reproduce the streamwise reduced distributions of the Bhatnagar--Gross--Krook (BGK) model and of DSMC to within 4\% up to Mach 12, and of DSMC to within 6\% at Mach 25, whereas two Maxwellians incur errors up to 50\% (BGK) and 37\% (DSMC). Collision histories identify the hard-sphere beam with unscattered stream molecules, which carry half of the peak viscous stress; the other populations are effective descriptions, not collision classes. Above Mach 5, Rankine--Hugoniot-scaled distributions are nearly Mach-independent, with corrections linear in $1/M^2$, consistent with the scaled kinetic equation. At equal normalised density, Mach 8 and 12 give the heat flux and stress at Mach 25 within 1\%, comparable to the accuracy of the reference; with one moment equation for position, physical-space profiles follow within 5\%. Comparing three prediction methods shows that the accuracy comes from the scaling; the Gaussian form adds compactness. Ahead of the shock, the mass of upstream-moving molecules decays as a stretched exponential with effective exponent 0.68--0.87. Collisions regenerate upstream-moving molecules, which explains a decay slower than free-flight survival; a linear cascade model with a downstream-Maxwellian source and no input from the shock interior reproduces the decay rates within 1.2\%.
Explore related subjects
Keep this discovery
Explore connections, maps & timelines
Ehsan Roohi. 2026-11-15. Population structure of the velocity distribution in normal shock waves: Gaussian representations, Mach-number similarity, and prediction. https://arxiv.science/abs/2611.20302
Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite arXiv Science by Indiciae: BibTeX · RIS.