pith. sign in

arxiv: 2406.09775 · v1 · submitted 2024-06-14 · 🧮 math.NA · cs.NA

A semi-implicit stochastic multiscale method for radiative heat transfer problem

Shan Zhang , Yajun Wang , Xiaofei Guan This is my paper

Pith reviewed 2026-05-24 00:05 UTC · model grok-4.3

classification 🧮 math.NA cs.NA
keywords radiative heat transferstochastic multiscale methodsemi-implicit schemecomposite materialsadditive noiseconvergence analysisnumerical experimentsmultiscale finite element
0
0 comments X

The pith

The semi-implicit stochastic multiscale method separates nonlinearity, high-dimensional randomness and spatial multiscale properties in radiative heat transfer so each can be treated independently.

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

The paper develops a numerical scheme for radiative heat transfer equations that contain additive noise and arise in composite materials. It first replaces the nonlinear radiation term at each time step with a semi-implicit predictor-corrector approximation, which produces a linear but still random and spatially multiscale equation. The random part is then reduced by expanding the noise in a complete orthogonal system and truncating the series, after which a spatial multiscale method built on efficient basis functions solves the resulting low-dimensional problem. Convergence analysis establishes an optimal rate for the combined procedure. A reader would care because the three sources of difficulty are handled one at a time rather than simultaneously, which opens the door to simulations that would otherwise be prohibitively expensive.

Core claim

The paper establishes that a semi-implicit predictor-corrected scheme first linearizes the strong nonlinearity, a complete orthogonal system then truncates the infinite-dimensional stochastic processes to a low-rank random equation, and spatial basis functions finally resolve the multiscale spatial structure, yielding a method whose convergence rate is optimal.

What carries the argument

The semi-implicit predictor-corrected scheme together with orthogonal truncation of the noise and spatial multiscale basis functions, which decouples the three computational difficulties so they can be addressed separately.

If this is right

  • The combined procedure attains the optimal convergence rate established by the analysis.
  • The method computes solutions for composite materials that possess a range of microstructures.
  • The separation into three independent stages permits each stage to be refined without redesigning the others.
  • Numerical experiments confirm both accuracy and computational efficiency on representative test problems.

Where Pith is reading between the lines

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

  • The same staged approach could be tested on other nonlinear stochastic partial differential equations that combine multiscale coefficients with random forcing.
  • If the orthogonal truncation is replaced by a data-driven low-rank representation of the noise, the random dimension might be reduced even further for specific material statistics.
  • The framework suggests that analogous decoupling might apply to time-dependent problems outside heat radiation whenever nonlinearity, randomness and spatial heterogeneity appear together.

Load-bearing premise

Truncating the infinite-dimensional stochastic processes with a complete orthogonal system captures the additive noise fluctuation without introducing errors large enough to change the overall solution behavior.

What would settle it

Numerical tests in which increasing the truncation dimension produces no further reduction in error, or in which the observed convergence rate falls below the predicted optimal rate, would show the separation strategy fails to deliver the claimed accuracy.

read the original abstract

In this paper, we propose and analyze a new semi-implicit stochastic multiscale method for the radiative heat transfer problem with additive noise fluctuation in composite materials. In the proposed method, the strong nonlinearity term induced by heat radiation is first approximated, by a semi-implicit predictor-corrected numerical scheme, for each fixed time step, resulting in a spatially random multiscale heat transfer equation. Then, the infinite-dimensional stochastic processes are modeled and truncated using a complete orthogonal system, facilitating the reduction of the model's dimensionality in the random space. The resulting low-rank random multiscale heat transfer equation is approximated and computed by using efficient spatial basis functions based multiscale method. The main advantage of the proposed method is that it separates the computational difficulty caused by the spatial multiscale properties, the high-dimensional randomness and the strong nonlinearity of the solution, so they can be overcome separately using different strategies. The convergence analysis is carried out, and the optimal rate of convergence is also obtained for the proposed semi-implicit stochastic multiscale method. Numerical experiments on several test problems for composite materials with various microstructures are also presented to gauge the efficiency and accuracy of the proposed semi-implicit stochastic multiscale method.

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

2 major / 2 minor

Summary. The paper proposes a semi-implicit stochastic multiscale method for the radiative heat transfer problem with additive noise in composite materials. Nonlinearity is handled via a semi-implicit predictor-corrector scheme per time step to yield a random multiscale equation; infinite-dimensional noise is truncated via a complete orthogonal system; the resulting low-dimensional random multiscale problem is discretized with spatial multiscale basis functions. Convergence analysis is claimed to deliver optimal rates, and numerical tests on composite microstructures are presented to demonstrate efficiency.

Significance. If the claimed separation of spatial multiscale, stochastic truncation, and nonlinearity difficulties is rigorously justified with optimal rates, the approach would provide a modular framework for handling coupled challenges in stochastic radiative transfer, which is relevant for modeling heat transfer in heterogeneous media. The explicit convergence result and numerical validation on varied microstructures would strengthen the contribution relative to purely deterministic multiscale methods.

major comments (2)
  1. [§4] §4 (Convergence analysis): The proof of the optimal convergence rate for the combined semi-implicit + truncation + multiscale scheme relies on an a priori bound for the truncation error of the stochastic expansion; this bound is not shown to be independent of the nonlinearity strength or the multiscale contrast, which is load-bearing for the claim that the three difficulties are overcome separately without rate degradation.
  2. [§3.2] §3.2 (Stochastic truncation): The choice of the orthogonal system and the truncation dimension M is presented without an explicit a posteriori error indicator or adaptive strategy; the analysis assumes the truncation error is absorbed into the optimal rate, but no numerical verification of this absorption is given for the highest-contrast microstructures.
minor comments (2)
  1. The notation for the semi-implicit predictor-corrector step (Eq. (3.5)–(3.7)) uses the same symbol for the intermediate and corrected solutions; a distinct superscript would improve readability.
  2. Figure 5 caption does not state the mesh size or the number of retained stochastic modes used in the plotted error curves.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and constructive comments on our manuscript. We address the major comments point by point below, providing clarifications and indicating planned revisions where appropriate.

read point-by-point responses

  1. Referee: [§4] §4 (Convergence analysis): The proof of the optimal convergence rate for the combined semi-implicit + truncation + multiscale scheme relies on an a priori bound for the truncation error of the stochastic expansion; this bound is not shown to be independent of the nonlinearity strength or the multiscale contrast, which is load-bearing for the claim that the three difficulties are overcome separately without rate degradation.

    Authors: We appreciate this observation. The truncation error bound is obtained from the L2-completeness of the orthogonal system together with the regularity assumptions on the additive noise; these are independent of nonlinearity strength because the semi-implicit predictor-corrector linearizes the radiation term at each time step before truncation. The multiscale contrast is likewise decoupled by the contrast-robust approximation properties of the spatial basis functions. The independence was implicit in the existing estimates rather than stated explicitly. We will add a short lemma or remark in Section 4 that isolates the truncation bound and confirms its independence from both parameters, thereby making the modular separation rigorous. revision: yes

  2. Referee: [§3.2] §3.2 (Stochastic truncation): The choice of the orthogonal system and the truncation dimension M is presented without an explicit a posteriori error indicator or adaptive strategy; the analysis assumes the truncation error is absorbed into the optimal rate, but no numerical verification of this absorption is given for the highest-contrast microstructures.

    Authors: The orthogonal system is selected for its spectral properties (completeness and rapid coefficient decay under the assumed smoothness of the noise), and M is chosen in practice by monitoring the decay of the expansion coefficients below a prescribed tolerance. While no a posteriori indicator is derived, the numerical experiments already cover several microstructures of varying contrast and the observed rates remain optimal. To strengthen the presentation we will insert a brief paragraph in Section 3.2 describing the coefficient-decay criterion and add one additional numerical table or figure that isolates the highest-contrast case to verify that the truncation contribution remains absorbed in the overall error. revision: partial

Circularity Check

0 steps flagged

No significant circularity; derivation uses standard sequential approximations

full rationale

The paper applies a semi-implicit predictor-corrector scheme to handle nonlinearity, followed by orthogonal truncation of stochastic processes and multiscale basis functions for spatial discretization. These steps are presented as independent standard techniques that separate the difficulties, with convergence analysis claimed to follow from the approximations. No self-definitional reductions, fitted inputs renamed as predictions, or load-bearing self-citations are evident in the abstract or description. The method is self-contained against external numerical analysis benchmarks without reducing to its own inputs by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

No free parameters, axioms, or invented entities are identifiable from the abstract alone; the method relies on standard numerical analysis techniques whose details are not provided.

pith-pipeline@v0.9.0 · 5738 in / 1017 out tokens · 22717 ms · 2026-05-24T00:05:49.284010+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

36 extracted references · 36 canonical work pages

  1. [1]

    Multiscale Modeling & Simulation 2, 124–157 (2003)

    Charrier, P., Dubroca, B.: Asymptotic transport models f or heat and mass transfer in reactive porous media. Multiscale Modeling & Simulation 2, 124–157 (2003)

    work page 2003

  2. [2]

    Computer Methods in Applied Mec hanics and Engineering 339, 298–319 (2018)

    Chung, E., Efendiev, Y., Leung, W.: Constraint energy min imizing generalized multi- scale finite element method. Computer Methods in Applied Mec hanics and Engineering 339, 298–319 (2018). Doi:https://doi.org/10.1016/j.cma.20 18.04.010

    work page doi:10.1016/j.cma.20 2018

  3. [3]

    Multiscale Modeling & Simulation 17, 1103–1122 (2019)

    Chung, E., Pun, S.: Online adaptive basis enrichment for m ixed cem- gmsfem. Multiscale Modeling & Simulation 17, 1103–1122 (2019). Doi:https://doi.org/10.1137/18M1222995

    work page doi:10.1137/18m1222995 2019

  4. [4]

    Applied Mathematics and Computation 218(23), 11496–11504 (2012)

    Cordero, A., Torregrosa, J.R., Vassileva, M.P.: Pseudoc omposition: a technique to design predictor–corrector methods for systems of nonlinear equa tions. Applied Mathematics and Computation 218(23), 11496–11504 (2012)

    work page 2012

  5. [5]

    Journal of the Society for Industri al and Applied Mathematics 11(1), 195–204 (1963)

    Douglas Jr, J., Jones Jr, B.: On predictor-corrector meth ods for nonlinear parabolic differential equations. Journal of the Society for Industri al and Applied Mathematics 11(1), 195–204 (1963)

    work page 1963

  6. [6]

    Journal of Computational Ph ysics 363, 39–54 (2018)

    Du, Q., Ju, L., Li, X., Qiao, Z.: Stabilized linear semi-im plicit schemes for the non- local cahn-hilliard equation. Journal of Computational Ph ysics 363, 39–54 (2018). Doi:https://doi.org/10.1016/j.jcp.2018.02.023 A semi-implicit stochastic multiscale method for radiativ e heat transfer problem 29

    work page doi:10.1016/j.jcp.2018.02.023 2018

  7. [7]

    Journal on Num erical Analysis 40, 1421– 1445 (2002)

    Du, Q., Zhang, T.: Numerical approximation of some linear stochastic partial differential equations driven by special additive noises. Journal on Num erical Analysis 40, 1421– 1445 (2002). Doi:https://doi.org/10.1137/S00361429013 87956

    work page doi:10.1137/s00361429013 2002

  8. [8]

    International Journal of Heat and Mass Transfer 16(1), 25– 40 (1973)

    Edwards, D., Balakrishnan, A.: Thermal radiation by comb ustion gases. International Journal of Heat and Mass Transfer 16(1), 25– 40 (1973). DOI https://doi.org/10.1016/0017-9310(73)90 248-2. URL https://www.sciencedirect.com/science/article/pii/0017931073902482

    work page doi:10.1016/0017-9310(73)90 1973

  9. [9]

    Journal of Computational Physics 251, 116–135 (2013)

    Efendiev, Y., Galvis, J., Hou, T.: Generalized multiscal e finite element meth- ods (GMsFEM). Journal of Computational Physics 251, 116–135 (2013). Doi:https://doi.org/10.1016/j.jcp.2013.04.045

    work page doi:10.1016/j.jcp.2013.04.045 2013

  10. [10]

    Internati onal Jour- nal for Multiscale Computational Engineering 12, 465–484 (2014)

    Efendiev, Y., Galvis, J., Li, G., Presho, M.: Generalize d multiscale fi- nite element methods: Oversampling strategies. Internati onal Jour- nal for Multiscale Computational Engineering 12, 465–484 (2014). Doi:https://doi.org/10.1615/IntJMultCompEng.2014007646

    work page doi:10.1615/intjmultcompeng.2014007646 2014

  11. [11]

    Journal of C omputational Physics 230(4), 937–955 (2011)

    Efendiev, Y., Galvis, J., W u, X.: Multiscale finite eleme nt methods for high-contrast problems using local spectral basis functions. Journal of C omputational Physics 230(4), 937–955 (2011)

    work page 2011

  12. [12]

    Journ al of Computational Physics 8(3), 313–342 (1971)

    Fleck Jr, J.A., Cummings Jr, J.: An implicit monte carlo s cheme for calculating time and frequency dependent nonlinear radiation transport. Journ al of Computational Physics 8(3), 313–342 (1971)

    work page 1971

  13. [13]

    Journal of Computational Physics 417, 109569 (2020)

    Fu, S., Chung, E., Mai, T.: Constraint energy minimizing generalized multiscale finite element method for nonlinear poroelasticity and elasticit y. Journal of Computational Physics 417, 109569 (2020)

    work page 2020

  14. [14]

    Multiscale Modeling & Simulation 18, 294–314 (2020)

    Heida, M., Kornhuber, R., Podlesny, J.: Fractal homogen ization of multiscale interface problems. Multiscale Modeling & Simulation 18, 294–314 (2020). Doi:https://doi.org/10.1137/18M1204759

    work page doi:10.1137/18m1204759 2020

  15. [15]

    Journal of Computational Physi cs 134, 169–189 (1997)

    Hou, T., W u, X.: A multiscale finite element method for ell iptic problems in composite materials and porous media. Journal of Computational Physi cs 134, 169–189 (1997). Doi:https://doi.org/10.1006/jcph.1997.5682

    work page doi:10.1006/jcph.1997.5682 1997

  16. [16]

    Mathematic s of computation 68, 913–943 (1999)

    Hou, T., W u, X., Cai, Z.: Convergence of a multiscale finit e element method for elliptic problems with rapidly oscillating coefficients. Mathematic s of computation 68, 913–943 (1999). Doi:https://doi.org/10.1090/S0025-5718-99-01 077-7

    work page doi:10.1090/s0025-5718-99-01 1999

  17. [17]

    Howell, J., MP., M., Daun, K., Siegel, R.: Thermal radiat ion heat transfer - Seventh edition (2020)

    work page 2020

  18. [18]

    Journal of computational physics 230, 3249–3263 (2011)

    Hu, G., Zegeling, P.A.: Simulating finger phenomena in po rous media with a mov- ing finite element method. Journal of computational physics 230, 3249–3263 (2011). Doi:https://doi.org/10.1016/j.jcp.2011.01.031

    work page doi:10.1016/j.jcp.2011.01.031 2011

  19. [19]

    Multiscale Modeling & Simulation 12, 694–724 (2014)

    Huang, J., Cao, L.: Global regularity and multiscale app roach for thermal ra- diation heat transfer. Multiscale Modeling & Simulation 12, 694–724 (2014). Doi:https://doi.org/10.1137/130919702

    work page doi:10.1137/130919702 2014

  20. [20]

    Applied Mathematics and Computation 266, 149– 168 (2015)

    Huang, J., Cao, L., Yang, C.: A multiscale algorithm for r adiative heat transfer equation with rapidly oscillating coefficients. Applied Mathematics and Computation 266, 149– 168 (2015). Doi:https://doi.org/10.1016/j.amc.2015.05 .048

    work page doi:10.1016/j.amc.2015.05 2015

  21. [21]

    Journal on Uncertainty Quantification 2, 397–422 (2014)

    Jagalur-Mohan, J., Sahni, O., Doostan, A., Oberai, A.: V ariational multiscale analysis: The fine-scale green’s function for stochastic partial diffe rential equations. Journal on Uncertainty Quantification 2, 397–422 (2014). Doi:https://doi.org/10.1137/13094035 9

    work page doi:10.1137/13094035 2014

  22. [22]

    International Jour nal of Numerical Analysis & Modeling (2013)

    Li, B., Sun, W.: Error analysis of linearized semi-impli cit galerkin finite element meth- ods for nonlinear parabolic equations. International Jour nal of Numerical Analysis & Modeling (2013)

    work page 2013

  23. [23]

    Journal on Scientific Computing 37, 103–133 (2015)

    Liao, Q., Willcox, K.: A domain decomposition approach f or uncer- tainty analysis. Journal on Scientific Computing 37, 103–133 (2015). Doi:https://doi.org/10.1137/140980508

    work page doi:10.1137/140980508 2015

  24. [24]

    Liu and Z

    Liu, Z., Qiao, Z.: Strong approximation of monotone stoc hastic partial different equa- tions driven by white noise. Journal of Numerical Analysis 40, 1074–1093 (2020). Doi:https://doi.org/10.1093/imanum/dry088

    work page doi:10.1093/imanum/dry088 2020

  25. [25]

    Communications i n Computational Physics 27, 1443–1469 (2020)

    Ma, C., Huang, J., Cao, L., Lin, Y.: Multiscale computati ons for the maxwell-schr¨ odinger system in heterogeneous nanostructures. Communications i n Computational Physics 27, 1443–1469 (2020). Doi:https://doi.org/10.4208/cicp.OA -2019-0004 30 Shan Zhang et al

    work page doi:10.4208/cicp.oa 2020

  26. [26]

    Journal of Compu tational Physics 227(16), 7561–7586 (2008)

    McClarren, R.G., Evans, T.M., Lowrie, R.B., Densmore, J .D.: Semi-implicit time inte- gration for pn thermal radiative transfer. Journal of Compu tational Physics 227(16), 7561–7586 (2008)

    work page 2008

  27. [27]

    Ap- plied Mathematics and Computation 188(2), 1587–1591 (2007)

    Noor, K.I., Noor, M.A.: Predictor–corrector halley met hod for nonlinear equations. Ap- plied Mathematics and Computation 188(2), 1587–1591 (2007)

    work page 2007

  28. [28]

    Multiscale Modeling & Simulat ion 16, 385–411 (2018)

    Ohlberger, M., Verfurth, B.: A new heterogeneous multis cale method for the helmholtz equation with high contrast. Multiscale Modeling & Simulat ion 16, 385–411 (2018). Doi:https://doi.org/10.1137/16M1108820

    work page doi:10.1137/16m1108820 2018

  29. [29]

    Journal of Scientific Computin g 37, 1–2 (2008)

    Schulze, T., Alexiades, V., Feng, X.: Multi-scale model ing and simulation in materials science, preface. Journal of Scientific Computin g 37, 1–2 (2008). Doi:https://doi.org/10.1007/s10915-008-9211-y

    work page doi:10.1007/s10915-008-9211-y 2008

  30. [30]

    Journal of Nonlinear Science 30, 23–66 (2020)

    W u, H., No’e, F.: Variational approach for learning mark ov processes from time series data. Journal of Nonlinear Science 30, 23–66 (2020). Doi:https://doi.org/10.1007/s00332-019-09567-y

    work page doi:10.1007/s00332-019-09567-y 2020

  31. [31]

    Com- puter Methods in Applied Mechanics and Engineering 356, 435–464 (2019)

    Yang, J., Mao, S., He, X., Yang, X., He, Y.: A diffuse interf ace model and semi-implicit energy stable finite element method for two-phase magnetohy drodynamic flows. Com- puter Methods in Applied Mechanics and Engineering 356, 435–464 (2019)

    work page 2019

  32. [32]

    Numerical Meth- ods for Partial Differential Equations: An International Jo urnal 32, 510–530 (2016)

    Yang, Z., Cui, J., W ang, Z., Zhang, Y.: Multiscale comput ational method for nonsta- tionary integrated heat transfer problem in periodic porou s materials. Numerical Meth- ods for Partial Differential Equations: An International Jo urnal 32, 510–530 (2016). Doi:https://doi.org/10.1002/num.22003

    work page doi:10.1002/num.22003 2016

  33. [33]

    in Proceedings of the 6th ASME-JSME Thermal Engineering Joint Conference pp

    Yuen, W.W.: Combined conductive/radiative heat transf er in high porosity fibrous in- sulation materials: Theory and experiment. in Proceedings of the 6th ASME-JSME Thermal Engineering Joint Conference pp. 16–20 (2003)

    work page 2003

  34. [34]

    Journal of Computational and Applied Mathematics 389, 913–943 (2021)

    Zhang, S., Guan, X., Jiang, L.: Convergence analysis of c onstraint energy minimizing generalized multiscale finite element method for a linear st ochastic parabolic partial differential equation driven by additive noises. Journal of Computational and Applied Mathematics 389, 913–943 (2021). Doi:https://doi.org/10.1016/j.cam.20 20.113328

    work page doi:10.1016/j.cam.20 2021

  35. [35]

    Zhang, T.: Numerical approximations of stochastic part ial differential equations. Ph.D. thesis, M. Phil thesis, Hong Kong University of Science and T echnology, Hong Kong (2000)

    work page 2000

  36. [36]

    Springer (2017)

    Zhang, Z., Karniadakis, G.E.: Numerical methods for sto chastic partial differential equa- tions with white noise. Springer (2017)

    work page 2017