pith. sign in

arxiv: 2606.25752 · v1 · pith:ZUOJACEYnew · submitted 2026-06-24 · ⚛️ physics.flu-dyn

Quantity-Dependent Bulk-to-Wall Observability of Surface Loading in Rarefied Hypersonic Flow over Triangular Protrusions

Elyas Lekzian , Ehsan Roohi This is my paper

Pith reviewed 2026-06-25 20:10 UTC · model grok-4.3

classification ⚛️ physics.flu-dyn
keywords rarefied hypersonic flowDSMCsurface loadingtriangular protrusionsbulk-to-wall dependenceinformation lengthpressure coefficientheat transfer coefficient
0
0 comments X

The pith

Rarefied wall loads on triangular protrusions draw from bulk gas at quantity-specific distances rather than any fixed neighborhood size.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper tests whether wall pressure, heat transfer, and shear on triangular protrusions in rarefied hypersonic flow can be predicted from local circular neighborhoods of increasing radius in the bulk flow. It trains surrogates on DSMC data across Mach 4-8 and Knudsen 0.1-0.8 and measures how much error remains when only partial neighborhoods are used. Full-domain information cuts Cp error from 45.5% to 13.8% and Cq error from 72.6% to 12.9%, while shear error falls only from 49.1% to 31.9%. R95, the smallest radius that nearly matches full-domain performance, varies by load type, with heat transfer showing the clearest dependence on distances of order the protrusion base length. The result indicates that no universal information length exists for rarefied surface loading.

Core claim

Rarefied surface loading has no single information length. Full-domain descriptors reduce errors from 45.5% to 13.8% for Cp and from 72.6% to 12.9% for Cq, whereas shear improves only from 49.1% to 31.9%. Heat transfer exhibits the clearest order-hs nonlocal support, pressure is frequently right-censored beyond 3hs, and shear saturates at shorter radii but remains least identifiable.

What carries the argument

R95, the smallest tested radius whose complete wall-profile error lies within 5% of the full-domain descriptor error, applied to circular neighborhoods summarized by weighted statistics, extrema, nearest-point values, and tangent-normal gradients.

If this is right

  • Pressure loading can often be approximated from neighborhoods no larger than 3hs while heat transfer requires neighborhoods at least hs in radius.
  • Shear stress remains the least predictable quantity even when the entire domain is included.
  • Coordinate-conditioned surrogates preserve the same quantity-dependent hierarchy of information lengths.
  • Closed-loop audit of the surrogates shows the largest preservation loss occurs for forward-facing heat transfer and shear.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Modeling codes that assume a single cutoff distance for all surface quantities will under-resolve heat transfer on protrusions.
  • Sensor placement strategies on hypersonic surfaces may need separate spacing rules for pressure versus heat-flux gauges.
  • The observed saturation of shear error suggests that local wall gradients alone are insufficient even at short range.

Load-bearing premise

Circular neighborhoods summarized by those particular statistics and gradients of velocity, temperature, and pressure capture the relevant bulk information needed to predict wall loads.

What would settle it

Re-running the error curves with a different neighborhood shape or with an entirely different set of bulk summary statistics and checking whether the hierarchy of R95 values across Cp, Cq, and shear disappears.

Figures

Figures reproduced from arXiv: 2606.25752 by Ehsan Roohi, Elyas Lekzian.

Figure 1
Figure 1. Figure 1: FIG. 1. Normalized triangular geometries used in this work; [PITH_FULL_IMAGE:figures/full_fig_p006_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2. Separation of the two computational roles in this study. Panel (a) summarizes the [PITH_FULL_IMAGE:figures/full_fig_p010_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3. Temperature validation for the Mach–rarefaction–orientation holdout: BWD protrusion, [PITH_FULL_IMAGE:figures/full_fig_p016_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4. Temperature validation for the height-interpolation holdout: FWD protrusion with [PITH_FULL_IMAGE:figures/full_fig_p016_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: FIG. 5. Temperature validation for the wall-temperature interpolation holdout: ISO protrusion [PITH_FULL_IMAGE:figures/full_fig_p016_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: FIG. 6. Streamwise-velocity validation for the three representative holdout cases. Row captions [PITH_FULL_IMAGE:figures/full_fig_p017_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: FIG. 7. Transverse-velocity validation for the same holdout cases. Velocity color scales are in m s [PITH_FULL_IMAGE:figures/full_fig_p018_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: FIG. 8. Pressure validation using the pressure-focused surrogate. Pressure color scales are in [PITH_FULL_IMAGE:figures/full_fig_p019_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: FIG. 9. Surface-only validation of protrusion-wall [PITH_FULL_IMAGE:figures/full_fig_p021_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: FIG. 10. Mean leave-one-physical-case-out (LOOCV) relative wall-profile error versus model input [PITH_FULL_IMAGE:figures/full_fig_p023_10.png] view at source ↗
Figure 12
Figure 12. Figure 12: FIG. 12. Closed-loop preservation of radius-dependent wall-load error. The vertically stacked [PITH_FULL_IMAGE:figures/full_fig_p028_12.png] view at source ↗
Figure 13
Figure 13. Figure 13: FIG. 13. Knudsen-number dependence of the lower-bound information horizon. Filled symbols [PITH_FULL_IMAGE:figures/full_fig_p030_13.png] view at source ↗
Figure 14
Figure 14. Figure 14: FIG. 14. Tree-bootstrap sensitivity of the lower-bound heat-transfer information horizon. Points [PITH_FULL_IMAGE:figures/full_fig_p033_14.png] view at source ↗
Figure 15
Figure 15. Figure 15: FIG. 15. Local-equilibrium entropy-rise proxy for the Mach–rarefaction–orientation validation [PITH_FULL_IMAGE:figures/full_fig_p035_15.png] view at source ↗
Figure 16
Figure 16. Figure 16: FIG. 16. Local-equilibrium entropy-rise proxy for height-interpolation validation cases. The com [PITH_FULL_IMAGE:figures/full_fig_p036_16.png] view at source ↗
read the original abstract

Localized protrusions on hypersonic vehicles generate pressure, heat-transfer, and shear loads whose rarefied response can depend on gas beyond the immediate wall neighborhood. This work quantifies that bulk-to-wall dependence for triangular protrusions and tests whether coordinate-conditioned surrogates preserve it. Geometry-consistent surrogates are trained for direct simulation Monte Carlo (DSMC) velocity, temperature, pressure, and wall-load profiles over Mach numbers 4--8, Knudsen numbers (Kn) 0.1--0.8, and three protrusion orientations. The central analysis is performed on raw DSMC fields. Around each wall point, circular neighborhoods of increasing radius are summarized by weighted statistics, extrema, nearest-point values, and tangent-normal gradients of velocity, temperature, and pressure. A fixed region-to-point diagnostic predicts the pressure coefficient ($C_p$), heat-transfer coefficient ($C_q$), and shear-stress magnitude ($|\tau|$). We define $R_{95}$ as the smallest tested radius whose complete wall-profile error lies within 5\% of the full-domain descriptor error. The principal physical result is that rarefied surface loading has no single information length. Full-domain descriptors reduce errors from 45.5\% to 13.8\% for $C_p$ and from 72.6\% to 12.9\% for $C_q$, whereas shear improves only from 49.1\% to 31.9\%. Heat transfer exhibits the clearest order-$h_s$ nonlocal support, where $h_s$ is the protrusion-base length. Pressure is frequently right-censored beyond $3h_s$, and shear saturates at shorter radii but remains least identifiable. Ridge-regression and threshold controls preserve this hierarchy, while a closed-loop audit shows partial surrogate preservation, with the largest degradation in forward-facing heat transfer and shear.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 0 minor

Summary. The manuscript analyzes bulk-to-wall information transfer for surface loads (Cp, Cq, and shear) on triangular protrusions in rarefied hypersonic flows (Mach 4-8, Kn 0.1-0.8) using raw DSMC fields. Circular neighborhoods of increasing radius around wall points are summarized via weighted statistics, extrema, nearest-point values, and tangent-normal gradients of velocity, temperature, and pressure; these descriptors are used to predict wall loads via a fixed region-to-point diagnostic. The central result is that rarefied surface loading has no single information length, quantified by R95 (the smallest radius where wall-profile error is within 5% of full-domain error), with full-domain descriptors yielding error reductions from 45.5% to 13.8% for Cp, 72.6% to 12.9% for Cq, and 49.1% to 31.9% for shear; heat transfer shows clearest order-hs nonlocal support while shear saturates earlier.

Significance. If the result holds, the quantity-specific nonlocal support demonstrated here would be a useful constraint for constructing reduced-order models or surrogates in rarefied hypersonic aerodynamics, showing that pressure and heat transfer require longer-range bulk information than shear. The direct use of raw DSMC fields, the explicit R95 metric, and the closed-loop surrogate audit provide concrete, reproducible diagnostics that could be extended to other geometries.

major comments (1)
  1. [Abstract (paragraph beginning 'Around each wall point, circular neighborhoods...')] Abstract (paragraph beginning 'Around each wall point, circular neighborhoods...'): The isotropic circular neighborhoods weight all azimuths equally when summarizing velocity, temperature, and pressure fields. At Mach 4-8 the molecular velocity distribution is strongly anisotropic, with molecules incident on a wall point originating predominantly from upstream; the reported hierarchy of error reductions (larger gains for Cp/Cq than for shear) and the claim of no single information length could therefore be an artifact of mixing irrelevant downstream data into the descriptors rather than a property of the underlying physics. A directional (e.g., upstream-weighted) neighborhood test is needed to confirm that the observed R95 values and differential improvements are robust.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the constructive comment on potential effects of flow anisotropy. We respond point-by-point below.

read point-by-point responses

  1. Referee: The isotropic circular neighborhoods weight all azimuths equally when summarizing velocity, temperature, and pressure fields. At Mach 4-8 the molecular velocity distribution is strongly anisotropic, with molecules incident on a wall point originating predominantly from upstream; the reported hierarchy of error reductions (larger gains for Cp/Cq than for shear) and the claim of no single information length could therefore be an artifact of mixing irrelevant downstream data into the descriptors rather than a property of the underlying physics. A directional (e.g., upstream-weighted) neighborhood test is needed to confirm that the observed R95 values and differential improvements are robust.

    Authors: The DSMC fields already incorporate the strong upstream bias of the molecular velocity distribution at Mach 4-8. Circular neighborhoods aggregate weighted statistics, extrema, and gradients over these inherently anisotropic fields without presupposing isotropy or a preferred direction; this choice identifies the minimal radius R95 while remaining geometry- and orientation-agnostic. The observed quantity-specific hierarchy (heat transfer requiring longest support, shear saturating earliest) is consistent with the distinct moments of the distribution function that govern each load. We agree that an explicit upstream-weighted test would provide additional confirmation. In revision we will add a supplementary directional-neighborhood analysis on representative cases to verify robustness of the R95 values and error reductions. revision: partial

Circularity Check

0 steps flagged

No circularity: empirical error analysis on raw DSMC fields

full rationale

The paper's central result—that rarefied surface loading has no single information length—is obtained by directly computing weighted statistics, extrema, nearest-point values, and gradients from raw DSMC velocity/temperature/pressure fields inside circular neighborhoods of increasing radius, then measuring prediction error for Cp, Cq, and |τ| against the actual DSMC wall loads. R95 is defined as the radius at which the wall-profile error reaches within 5% of the full-domain error; this reduction is measured, not fitted by construction. No self-definitional steps, no fitted parameters renamed as predictions, and no load-bearing self-citations appear in the derivation. The analysis remains self-contained against the external DSMC benchmark.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Review performed on abstract only; no explicit free parameters, axioms, or invented entities are stated in the provided text.

axioms (1)
  • domain assumption DSMC simulations accurately represent the rarefied gas dynamics for the stated Mach and Knudsen ranges
    The entire analysis rests on DSMC-generated velocity, temperature, and pressure fields.

pith-pipeline@v0.9.1-grok · 5880 in / 1331 out tokens · 37661 ms · 2026-06-25T20:10:09.844945+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

58 extracted references · 58 canonical work pages

  1. [1]

    P., ``Summary of hypersonic boundary-layer transition experiments on blunt bodies with roughness", Journal of Spacecraft and Rockets, 45, 1090--1105 (2008)

    Schneider, S. P., ``Summary of hypersonic boundary-layer transition experiments on blunt bodies with roughness", Journal of Spacecraft and Rockets, 45, 1090--1105 (2008). doi:10.2514/1.37431 https://doi.org/10.2514/1.37431

    work page doi:10.2514/1.37431 2008

  2. [2]

    A., Horvath, T

    Berry, S. A., Horvath, T. J., ``Discrete-roughness transition for hypersonic flight vehicles", Journal of Spacecraft and Rockets, 45, 216--227 (2008). doi:10.2514/1.30970 https://doi.org/10.2514/1.30970

    work page doi:10.2514/1.30970 2008

  3. [3]

    Grotowsky, I. M. G., Ballmann, J., ``Numerical investigation of hypersonic step-flows", Shock Waves, 10, 57--72 (2000). doi:10.1007/s001930050179 https://doi.org/10.1007/s001930050179

    work page doi:10.1007/s001930050179 2000

  4. [4]

    G., Stollery, J

    Estruch, D., MacManus, D. G., Stollery, J. L., Lawson, N. J., Garry, K. P., ``Hypersonic interference heating in the vicinity of surface protuberances", Experiments in Fluids, 49, 683--699 (2010). doi:10.1007/s00348-010-0844-x https://doi.org/10.1007/s00348-010-0844-x

    work page doi:10.1007/s00348-010-0844-x 2010

  5. [5]

    AIAA 2010-1572 (2010)

    Chang, C.-L., Choudhari, M., Li, F., ``Numerical computations of hypersonic boundary-layer over surface irregularities" in 48th AIAA Aerospace Sciences Meeting, pp. AIAA 2010-1572 (2010). doi:10.2514/6.2010-1572 https://doi.org/10.2514/6.2010-1572

    work page doi:10.2514/6.2010-1572 2010

  6. [6]

    doi:10.1080/19942060.2012.11015424 https://doi.org/10.1080/19942060.2012.11015424

    Qamar, A., Sanghi, S., ``Aerodynamics characteristic of axisymmetric surface protuberance in supersonic regime", Engineering Applications of Computational Fluid Mechanics, 6, 316--335 (2012). doi:10.1080/19942060.2012.11015424 https://doi.org/10.1080/19942060.2012.11015424

    work page doi:10.1080/19942060.2012.11015424 2012

  7. [7]

    D., Theofilis, V., ``Laminar-turbulent transition induced by a discrete roughness element in a supersonic boundary layer", Journal of Fluid Mechanics, 735, 613--646 (2013)

    De Tullio, N., Paredes, P., Sandham, N. D., Theofilis, V., ``Laminar-turbulent transition induced by a discrete roughness element in a supersonic boundary layer", Journal of Fluid Mechanics, 735, 613--646 (2013). doi:10.1017/jfm.2013.520 https://doi.org/10.1017/jfm.2013.520

    work page doi:10.1017/jfm.2013.520 2013

  8. [8]

    V., Frendi, A., ``Interaction of three-dimensional protuberances with a supersonic turbulent boundary layer", AIAA Journal, 51, 1657--1666 (2013)

    Hahn, P. V., Frendi, A., ``Interaction of three-dimensional protuberances with a supersonic turbulent boundary layer", AIAA Journal, 51, 1657--1666 (2013). doi:10.2514/1.J052101 https://doi.org/10.2514/1.J052101

    work page doi:10.2514/1.j052101 2013

  9. [9]

    S., Reddy, K

    Kumar, C. S., Reddy, K. P. J., ``Experimental investigation of heat fluxes in the vicinity of protuberances on a flat plate at hypersonic speeds", Journal of Heat Transfer, 135, 121701 (2013). doi:10.1115/1.4024667 https://doi.org/10.1115/1.4024667

    work page doi:10.1115/1.4024667 2013

  10. [10]

    S., Reddy, K

    Kumar, C. S., Reddy, K. P. J., ``Hypersonic interference heating on cones with short three-dimensional protuberances", Experimental Thermal and Fluid Science, 55, 29--41 (2014). doi:10.1016/j.expthermflusci.2014.02.013 https://doi.org/10.1016/j.expthermflusci.2014.02.013

    work page doi:10.1016/j.expthermflusci.2014.02.013 2014

  11. [11]

    Avallone, F., Schrijer, F. F. J., Cardone, G., ``Infrared thermography of transition due to isolated roughness elements in hypersonic flows", Physics of Fluids, 28, 024106 (2016). doi:10.1063/1.4941527 https://doi.org/10.1063/1.4941527

    work page doi:10.1063/1.4941527 2016

  12. [12]

    doi:10.1016/j.compfluid.2017.08.006 https://doi.org/10.1016/j.compfluid.2017.08.006

    Duan, Z., Xiao, Z., ``Hypersonic transition induced by three isolated roughness elements on a flat plate", Computers and Fluids, 157, 1--13 (2017). doi:10.1016/j.compfluid.2017.08.006 https://doi.org/10.1016/j.compfluid.2017.08.006

    work page doi:10.1016/j.compfluid.2017.08.006 2017

  13. [13]

    doi:10.1017/jfm.2018.706 https://doi.org/10.1017/jfm.2018.706

    Di Giovanni, A., Stemmer, C., ``Cross-flow-type breakdown induced by distributed roughness in the boundary layer of a hypersonic capsule configuration", Journal of Fluid Mechanics, 856, 470--503 (2018). doi:10.1017/jfm.2018.706 https://doi.org/10.1017/jfm.2018.706

    work page doi:10.1017/jfm.2018.706 2018

  14. [14]

    R., ``Analysis of hypersonic flow over pin protrusions on a blunt body" in Recent Advances in Sustainable Technologies, pp

    Akshay, N., Nagaraja, S. R., ``Analysis of hypersonic flow over pin protrusions on a blunt body" in Recent Advances in Sustainable Technologies, pp. 293--302 (2021). doi:10.1007/978-981-16-0976-3\_28 https://doi.org/10.1007/978-981-16-0976-3_28

    work page doi:10.1007/978-981-16-0976-3 2021

  15. [15]

    B., ``Wall roughness effects on the supersonic flow over a circular cylinder", Shock Waves, 32, 643--663 (2022)

    Blanco-Casares, A., Jacobs, G. B., ``Wall roughness effects on the supersonic flow over a circular cylinder", Shock Waves, 32, 643--663 (2022). doi:10.1007/s00193-022-01098-y https://doi.org/10.1007/s00193-022-01098-y

    work page doi:10.1007/s00193-022-01098-y 2022

  16. [16]

    doi:10.2514/1.A35319 https://doi.org/10.2514/1.A35319

    Tsutsui, F., Takagi, Y., Takimoto, H., Kitamura, K., Nonaka, S., ``Side force characteristics of slender-bodied supersonic vehicle with two protuberances", Journal of Spacecraft and Rockets, 59, 1697--1712 (2022). doi:10.2514/1.A35319 https://doi.org/10.2514/1.A35319

    work page doi:10.2514/1.a35319 2022

  17. [17]

    Chandrakumar, V. C. M., Nagaraja, S. R., ``Study of effect of protrusions on cones in hypersonic flow" in Recent Developments in Mechanics and Design, pp. 235--242 (2023). doi:10.1007/978-981-19-4140-5\_20 https://doi.org/10.1007/978-981-19-4140-5_20

    work page doi:10.1007/978-981-19-4140-5 2023

  18. [18]

    doi:10.1063/5.0137902 https://doi.org/10.1063/5.0137902

    Haley, C., Zhong, X., ``Roughness effect on hypersonic second mode instability and transition on a cone", Physics of Fluids, 35, 014108 (2023). doi:10.1063/5.0137902 https://doi.org/10.1063/5.0137902

    work page doi:10.1063/5.0137902 2023

  19. [19]

    doi:10.2514/1.A35908 https://doi.org/10.2514/1.A35908

    Leopardi, M., Paciorri, R., Assonitis, A., Neri, A., Barbagallo, D., ``Effects of protuberances on surface loads on the Vega-C launch vehicle", Journal of Spacecraft and Rockets (2024). doi:10.2514/1.A35908 https://doi.org/10.2514/1.A35908

    work page doi:10.2514/1.a35908 2024

  20. [20]

    I., Harvey, J

    Pullin, D. I., Harvey, J. K., ``Direct simulation calculations of the rarefied flow past a forward-facing step", AIAA Journal, 15, 124--126 (1977). doi:10.2514/3.7310 https://doi.org/10.2514/3.7310

    work page doi:10.2514/3.7310 1977

  21. [21]

    Leite, P. H. M., Santos, W. F. N., ``Computational analysis of a rarefied hypersonic flow over combined gap/step geometries", Advances in Aerospace Sciences, 7, 369--394 (2015). doi:10.1051/eucass/201507369 https://doi.org/10.1051/eucass/201507369

    work page doi:10.1051/eucass/201507369 2015

  22. [22]

    C., ``DSMC simulation of rarefied gas flow over a wall mounted cube" in ASME Fluids Engineering Division Summer Meeting (2019)

    Nabapure, D., Murthy, R. C., ``DSMC simulation of rarefied gas flow over a wall mounted cube" in ASME Fluids Engineering Division Summer Meeting (2019). doi:10.1115/AJKFluids2019-5447 https://doi.org/10.1115/AJKFluids2019-5447

    work page doi:10.1115/ajkfluids2019-5447 2019

  23. [23]

    doi:10.1088/1757-899X/887/1/012015 https://doi.org/10.1088/1757-899X/887/1/012015

    Wang, L., Fang, S., ``Effect of obstacle's length-to-height ratio on aerodynamic quantities of rarefied hypersonic flow", IOP Conference Series: Materials Science and Engineering, 887, 012015 (2020). doi:10.1088/1757-899X/887/1/012015 https://doi.org/10.1088/1757-899X/887/1/012015

    work page doi:10.1088/1757-899x/887/1/012015 2020

  24. [24]

    C., ``DSMC investigation of rarefied gas flow over a 2D forward-facing step: Effect of Knudsen number", Acta Astronautica, 178, 89--109 (2021)

    Nabapure, D., Murthy, R. C., ``DSMC investigation of rarefied gas flow over a 2D forward-facing step: Effect of Knudsen number", Acta Astronautica, 178, 89--109 (2021). doi:10.1016/j.actaastro.2020.08.030 https://doi.org/10.1016/j.actaastro.2020.08.030

    work page doi:10.1016/j.actaastro.2020.08.030 2021

  25. [25]

    doi:10.1017/jfm.2020.73 https://doi.org/10.1017/jfm.2020.73

    Manela, A., Gibelli, L., ``Free-molecular and near-free-molecular gas flows over backward facing steps", Journal of Fluid Mechanics, 889, A22 (2020). doi:10.1017/jfm.2020.73 https://doi.org/10.1017/jfm.2020.73

    work page doi:10.1017/jfm.2020.73 2020

  26. [26]

    doi:10.1016/j.vacuum.2018.06.026 https://doi.org/10.1016/j.vacuum.2018.06.026

    Gavasane, A., Agrawal, A., Bhandarkar, U., ``Study of rarefied gas flows in backward facing micro-step using direct simulation Monte Carlo", Vacuum, 155, 249--259 (2018). doi:10.1016/j.vacuum.2018.06.026 https://doi.org/10.1016/j.vacuum.2018.06.026

    work page doi:10.1016/j.vacuum.2018.06.026 2018

  27. [27]

    doi:10.1063/5.0090538 https://doi.org/10.1063/5.0090538

    Mahdavi, Amirmehran, Roohi, Ehsan, ``A study on micro-step flow using a hybrid direct simulation Monte Carlo-Fokker-Planck approach", Physics of Fluids, 34, 062007 (2022). doi:10.1063/5.0090538 https://doi.org/10.1063/5.0090538

    work page doi:10.1063/5.0090538 2022

  28. [28]

    doi:10.1016/j.applthermaleng.2025.126763 https://doi.org/10.1016/j.applthermaleng.2025.126763

    Sabouri, Moslem, Lekzian, Elyas, ``Thermal and flow effects of corner rounding in rarefied hypersonic step flows", Applied Thermal Engineering, 274, 126763 (2025). doi:10.1016/j.applthermaleng.2025.126763 https://doi.org/10.1016/j.applthermaleng.2025.126763

    work page doi:10.1016/j.applthermaleng.2025.126763 2025

  29. [29]

    doi:10.1007/s10494-026-00741-3 https://doi.org/10.1007/s10494-026-00741-3

    Sabouri, Moslem, Lekzian, Elyas, ``Study of Triangle-shaped Protrusions Exposed to High-speed Flow in Rarefied Regime", Flow, Turbulence and Combustion, 116, 47 (2026). doi:10.1007/s10494-026-00741-3 https://doi.org/10.1007/s10494-026-00741-3

    work page doi:10.1007/s10494-026-00741-3 2026

  30. [30]

    doi:10.1017/9781139683494 https://doi.org/10.1017/9781139683494

    Boyd, Iain D., Schwartzentruber, Thomas E., Nonequilibrium Gas Dynamics and Molecular Simulation, Cambridge University Press (2017). doi:10.1017/9781139683494 https://doi.org/10.1017/9781139683494

    work page doi:10.1017/9781139683494 2017

  31. [31]

    doi:10.1063/1.5099042 https://doi.org/10.1063/1.5099042

    Stefanov, Stefan K., ``On the basic concepts of the direct simulation Monte Carlo method", Physics of Fluids, 31, 067104 (2019). doi:10.1063/1.5099042 https://doi.org/10.1063/1.5099042

    work page doi:10.1063/1.5099042 2019

  32. [32]

    J., Moore, S

    Plimpton, S. J., Moore, S. G., Borner, A., Stagg, A. K., Koehler, T. P., Torczynski, J. R., Gallis, M. A., ``Direct simulation Monte Carlo on petaflop supercomputers and beyond", Physics of Fluids, 31, 086101 (2019). doi:10.1063/1.5108534 https://doi.org/10.1063/1.5108534

    work page doi:10.1063/1.5108534 2019

  33. [33]

    doi:10.2514/3.49461 https://doi.org/10.2514/3.49461

    Becker, M., Robben, F., Cattolica, R., ``Velocity distribution functions near the leading edge of a flat plate", AIAA Journal, 12, 1247--1253 (1974). doi:10.2514/3.49461 https://doi.org/10.2514/3.49461

    work page doi:10.2514/3.49461 1974

  34. [34]

    doi:10.1016/j.compfluid.2011.04.019 https://doi.org/10.1016/j.compfluid.2011.04.019

    Sun, Z.-X., Tang, Z., He, Y.-L., Tao, W.-Q., ``Proper cell dimension and number of particles per cell for DSMC", Computers and Fluids, 50, 1--9 (2011). doi:10.1016/j.compfluid.2011.04.019 https://doi.org/10.1016/j.compfluid.2011.04.019

    work page doi:10.1016/j.compfluid.2011.04.019 2011

  35. [35]

    doi:10.1016/j.compfluid.2019.03.007 https://doi.org/10.1016/j.compfluid.2019.03.007

    Shamseddine, M., Lakkis, I., ``A novel spatio-temporally adaptive parallel three-dimensional DSMC solver for unsteady rarefied micro/nano gas flows", Computers and Fluids, 186, 1--14 (2019). doi:10.1016/j.compfluid.2019.03.007 https://doi.org/10.1016/j.compfluid.2019.03.007

    work page doi:10.1016/j.compfluid.2019.03.007 2019

  36. [36]

    doi:10.1088/1402-4896/ad5a46 https://doi.org/10.1088/1402-4896/ad5a46

    Sabouri, Moslem, Zakeri, R., Ebrahimi, A., ``Improving computational efficiency in DSMC simulations of vacuum gas dynamics with a fixed number of particles per cell", Physica Scripta, 99, 085213 (2024). doi:10.1088/1402-4896/ad5a46 https://doi.org/10.1088/1402-4896/ad5a46

    work page doi:10.1088/1402-4896/ad5a46 2024

  37. [37]

    Raissi, M., Perdikaris, P., Karniadakis, G. E., ``Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations", Journal of Computational Physics, 378, 686--707 (2019). doi:10.1016/j.jcp.2018.10.045 https://doi.org/10.1016/j.jcp.2018.10.045

    work page doi:10.1016/j.jcp.2018.10.045 2019

  38. [38]

    Kevrekidis, Lu Lu, Paris Perdikaris, Sifan Wang, and Liu Yang

    Karniadakis, G. E., Kevrekidis, I. G., Lu, L., Perdikaris, P., Wang, S., Yang, L., ``Physics-informed machine learning", Nature Reviews Physics, 3, 422--440 (2021). doi:10.1038/s42254-021-00314-5 https://doi.org/10.1038/s42254-021-00314-5

    work page doi:10.1038/s42254-021-00314-5 2021

  39. [39]

    Mosaic: A Benchmark Suite for Differentiable Physics Solvers — Rehmann et al., 2026 13

    Lu, Lu, Jin, Pengzhan, Pang, Guofei, Zhang, Zhongqiang, Karniadakis, George Em, ``Learning nonlinear operators via DeepONet based on the universal approximation theorem of operators", Nature Machine Intelligence, 3, 218--229 (2021). doi:10.1038/s42256-021-00302-5 https://doi.org/10.1038/s42256-021-00302-5

    work page doi:10.1038/s42256-021-00302-5 2021

  40. [40]

    Probabilistic neural networks for fluid flow surrogate modeling and data recovery

    Maulik, Romit, Fukami, Kai, Ramachandra, Nesar, Fukagata, Koji, Taira, Kunihiko, ``Probabilistic neural networks for fluid flow surrogate modeling and data recovery", Physical Review Fluids, 5, 104401 (2020). doi:10.1103/PhysRevFluids.5.104401 https://doi.org/10.1103/PhysRevFluids.5.104401

    work page doi:10.1103/physrevfluids.5.104401 2020

  41. [41]

    Assessments of epistemic uncertainty using Gaussian stochastic weight averaging for fluid-flow regression

    Morimoto, Masaki, Fukami, Kai, Maulik, Romit, Vinuesa, Ricardo, Fukagata, Koji, ``Assessments of epistemic uncertainty using Gaussian stochastic weight averaging for fluid-flow regression", Physica D, 440, 133454 (2022). doi:10.1016/j.physd.2022.133454 https://doi.org/10.1016/j.physd.2022.133454

    work page doi:10.1016/j.physd.2022.133454 2022

  42. [42]

    doi:10.1016/j.jcp.2025.114432 https://doi.org/10.1016/j.jcp.2025.114432

    Peyvan, Ahmad, Kumar, Varun, Karniadakis, George Em, ``Fusion-DeepONet: a data-efficient neural operator for geometry-dependent hypersonic and supersonic flows", Journal of Computational Physics, 544, 114432 (2026). doi:10.1016/j.jcp.2025.114432 https://doi.org/10.1016/j.jcp.2025.114432

    work page doi:10.1016/j.jcp.2025.114432 2026

  43. [43]

    doi:10.1007/s10404-026-02899-8 https://doi.org/10.1007/s10404-026-02899-8

    Roohi, Ehsan, Mahdavi, Amirmehran, ``Analysis of the rarefied flow at micro-step using a DeepONet surrogate model with a physics-guided zonal loss function", Microfluidics and Nanofluidics, 30, 44 (2026). doi:10.1007/s10404-026-02899-8 https://doi.org/10.1007/s10404-026-02899-8

    work page doi:10.1007/s10404-026-02899-8 2026

  44. [44]

    Z. Luo, L. Wang, J. Xu, M. Chen, J. Yuan, and A. C. C. Tan, ``Flow reconstruction from sparse sensors based on reduced-order autoencoder state estimation,'' Physics of Fluids 35, 077107 (2023). doi:10.1063/5.0155039 https://doi.org/10.1063/5.0155039

    work page doi:10.1063/5.0155039 2023

  45. [45]

    M. Y. Hosseini and Y. Shiri, ``Flow field reconstruction from sparse sensor measurements with physics-informed neural networks,'' Physics of Fluids 36, 073606 (2024). doi:10.1063/5.0211680 https://doi.org/10.1063/5.0211680

    work page doi:10.1063/5.0211680 2024

  46. [46]

    Fukami, R

    K. Fukami, R. Maulik, N. Ramachandra, K. Fukagata, and K. Taira, ``Global field reconstruction from sparse sensors with Voronoi tessellation-assisted deep learning,'' Nature Machine Intelligence 3, 945--951 (2021). doi:10.1038/s42256-021-00402-2 https://doi.org/10.1038/s42256-021-00402-2

    work page doi:10.1038/s42256-021-00402-2 2021

  47. [47]

    J. E. Santos, Z. R. Fox, A. Mohan, D. O'Malley, H. Viswanathan, and N. Lubbers, ``Development of the Senseiver for efficient field reconstruction from sparse observations,'' Nature Machine Intelligence 5, 1317--1325 (2023). doi:10.1038/s42256-023-00746-x https://doi.org/10.1038/s42256-023-00746-x

    work page doi:10.1038/s42256-023-00746-x 2023

  48. [48]

    X. Zhao, X. Chen, Z. Gong, W. Zhou, W. Yao, and Y. Zhang, ``RecFNO: A resolution-invariant flow and heat field reconstruction method from sparse observations via Fourier neural operator,'' International Journal of Thermal Sciences 195, 108619 (2024). doi:10.1016/j.ijthermalsci.2023.108619 https://doi.org/10.1016/j.ijthermalsci.2023.108619

    work page doi:10.1016/j.ijthermalsci.2023.108619 2024

  49. [49]

    J. P. Williams, O. Zahn, and J. N. Kutz, ``Sensing with shallow recurrent decoder networks,'' Proceedings of the Royal Society A 480, 20240054 (2024). doi:10.1098/rspa.2024.0054 https://doi.org/10.1098/rspa.2024.0054

    work page doi:10.1098/rspa.2024.0054 2024

  50. [50]

    X. Kong, Q. Fan, J. Li, and W. He, ``Reconstructing flow field from sparse sensor data: A deep learning framework combining autoencoder and cross-attention mechanism,'' Physics of Fluids 37, 097119 (2025). doi:10.1063/5.0289122 https://doi.org/10.1063/5.0289122

    work page doi:10.1063/5.0289122 2025

  51. [51]

    X. Wang, Z. Li, Y. Zhang, H. Wen, and B. Wang, ``Observation-guided generative flow field reconstruction of rotating detonation combustor,'' Physics of Fluids 37, 116110 (2025). doi:10.1063/5.0300394 https://doi.org/10.1063/5.0300394

    work page doi:10.1063/5.0300394 2025

  52. [52]

    Zhang, J.-Z

    Y.-Q. Zhang, J.-Z. Peng, Z.-Q. Wang, W.-J. Yuan, Y.-B. Li, Z.-F. Zhou, H.-R. Xie, and W.-T. Wu, ``High-fidelity flow-field reconstruction from sparse sensors using a hybrid graph neural network-transformer architecture,'' Physics of Fluids 38, 023610 (2026). doi:10.1063/5.0307325 https://doi.org/10.1063/5.0307325

    work page doi:10.1063/5.0307325 2026

  53. [53]

    Manohar, B

    K. Manohar, B. W. Brunton, J. N. Kutz, and S. L. Brunton, ``Data-driven sparse sensor placement for reconstruction: Demonstrating the benefits of exploiting known patterns,'' IEEE Control Systems Magazine 38, 63--86 (2018). doi:10.1109/MCS.2018.2810460 https://doi.org/10.1109/MCS.2018.2810460

    work page doi:10.1109/mcs.2018.2810460 2018

  54. [54]

    N. B. Erichson, L. Mathelin, Z. Yao, S. L. Brunton, M. W. Mahoney, and J. N. Kutz, ``Shallow neural networks for fluid flow reconstruction with limited sensors,'' Proceedings of the Royal Society A 476, 20200097 (2020). doi:10.1098/rspa.2020.0097 https://doi.org/10.1098/rspa.2020.0097

    work page doi:10.1098/rspa.2020.0097 2020

  55. [55]

    Liang, C

    Y. Liang, C. Hou, G. Y. Cornejo Maceda, A. Ianiro, S. Discetti, A. Meil\'an-Vila, D. Sornette, S. C. Lera, J. Chen, X. He, and B. R. Noack, ``Sensor optimization for urban wind estimation with a cluster-based probabilistic framework,'' Physics of Fluids 38, 045131 (2026). doi:10.1063/5.0303585 https://doi.org/10.1063/5.0303585

    work page doi:10.1063/5.0303585 2026

  56. [56]

    X. Li, B. Hu, and L. Wu, ``A non-localized spatial--temporal constitutive relation in rarefied gas dynamics,'' Physics of Fluids 36, 102018 (2024). doi:10.1063/5.0228567 https://doi.org/10.1063/5.0228567

    work page doi:10.1063/5.0228567 2024

  57. [57]

    Geurts, D

    P. Geurts, D. Ernst, and L. Wehenkel, ``Extremely randomized trees,'' Machine Learning 63, 3--42 (2006). doi:10.1007/s10994-006-6226-1 https://doi.org/10.1007/s10994-006-6226-1

    work page doi:10.1007/s10994-006-6226-1 2006

  58. [58]

    The Annals of Statistics7(1), 1 – 26 (1979).https://doi.org/10.1214/aos/1176344552,https: //doi.org/10.1214/aos/11763445525

    B. Efron, ``Bootstrap methods: Another look at the jackknife,'' The Annals of Statistics 7, 1--26 (1979). doi:10.1214/aos/1176344552 https://doi.org/10.1214/aos/1176344552

    work page doi:10.1214/aos/1176344552 1979