Pith. sign in

REVIEW 3 major objections 5 minor 108 references

This review argues that physics-informed neural networks, with conservation laws written into their loss functions, can deliver mesh-free, data-efficient models that match CFD accuracy for flame dynamics, emissions, and instabilities at a f

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

T0 review · deepseek-v4-flash

2026-08-04 21:17 UTC pith:QH3W2H7R

load-bearing objection A useful but overhyped review: the survey of PINNs for combustion is worth reading, but the aerospace-propulsion claim outruns the cited evidence. the 3 major comments →

arxiv 2509.08094 v1 pith:QH3W2H7R submitted 2025-09-09 physics.flu-dyn physics.chem-ph

Physics-Informed Neural Networks in Clean Combustion: A Pathway to Sustainable Aerospace Propulsion

Mahmood Mousavi , Caleb Caldwell , Jacob Baltes , Muteb Aljasem , Bok Jik Lee This is my paper
classification physics.flu-dyn physics.chem-ph
keywords Physics-informed neural networksclean combustionaerospace propulsionflame dynamicsturbulent combustionNOx emissionsthermoacoustic instabilitiesstiff chemical kinetics
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The paper is a review whose central argument is that physics-informed neural networks (PINNs) can accelerate and sometimes replace conventional computational fluid dynamics in clean combustion, with aerospace propulsion as the target application. By embedding conservation equations directly into the training loss, PINNs are claimed to achieve accurate, mesh-free modeling even when data are sparse, while reducing computational cost. The concrete evidence cited includes 6.0-14.6x speedups for chemical source-term integration, 2.3-4.9x overall reactive-flow simulation speedups, and test R-squared values around 0.96 for NOx emission prediction. The reason this matters is that these capabilities would enable real-time emission control, instability management, and faster design exploration for cleaner engines. The review is forward-looking, but its engine-level conclusions rest on results obtained from canonical and lab-scale cases rather than full-scale propulsion demonstrations.

Core claim

The paper's thesis is that PINNs, by adding PDE residuals and conservation constraints to a neural network's loss function, can solve forward and inverse combustion problems with less data and lower cost than mesh-based CFD. It organizes the evidence across flame dynamics, turbulent combustion, emissions prediction, and thermoacoustic instabilities. The strongest quantitative anchors are the CRK-PINN results for stiff reaction kinetics, the monotonicity-constrained PINN results for NOx from coal-fired boilers, and the thermoacoustic PINN reconstruction of acoustic fields from sparse sensor data. The review claims these examples show a pathway to next-generation aerospace engines where PINN-b

What carries the argument

The composite loss function is the central mechanism: it sums data-misfit terms with weighted residuals of the governing PDEs, boundary and initial conditions, and additional physical constraints such as enthalpy conservation, element conservation, mole-fraction conservation, monotonicity, and derivative constraints. Nested inside this machinery are specialized constructions—CRK-PINNs with logarithmic normalization for stiff kinetics, DPINNs with derivative-constrained losses for steep-gradient reacting flows, and DeepONet-style architectures for parametric turbulent closures. The loss function is what lets the network trade data fidelity against physical consistency, which is the source of

Load-bearing premise

That speed and accuracy measured on canonical and lab-scale combustion cases—0-D autoignition, one-dimensional laminar flames, lab-scale bluff-body combustors, and coal-boiler NOx data—transfer to full-scale, three-dimensional, high-Reynolds-number aerospace propulsion with stiff multi-species kinetics and noisy sensors.

What would settle it

Run a trained PINN surrogate, such as a CRK-PINN chemistry model or a thermoacoustic PINN, against high-fidelity LES or experimental data for a full-scale annular or rocket combustor across off-design operating conditions; if the wall-clock speedup over the conventional solver disappears, or prediction error grows well beyond the lab-case R-squared around 0.96 once operating points leave the training distribution, the claimed aerospace pathway is not supported.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

Share X Bluesky LinkedIn Reddit HN

If this is right

  • Stiff chemical source terms can be computed 6.0-14.6 times faster than direct integration, and overall reactive-flow simulations can run 2.3-4.9 times faster, making detailed kinetics practical in iterative design and control loops.
  • Physics-constrained NOx models with monotonicity constraints reach test R-squared values around 0.96-0.97, outperforming plain neural networks, random forests, and support-vector regression under varying operating conditions where labeled data are scarce.
  • Sparse acoustic pressure and heat-release measurements can be converted into complete spatiotemporal thermoacoustic fields, enabling instability diagnosis and parameter estimation without dense sensor arrays.
  • Hybrid frameworks that keep CFD for the flow field and use PINNs only for chemical source terms or subgrid closures can cut stiff-chemistry overhead while preserving the validation pedigree of conventional solvers.
  • Because trained PINNs evaluate through forward passes, they can serve as fast predictive models in model predictive control, holding emissions and efficiency constraints in real time.

Where Pith is reading between the lines

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

  • The engine-level claim in the title and conclusion is an extrapolation: the speedup and accuracy numbers come from 0-D autoignition problems, 1-D laminar flames, lab-scale bluff-body and turbulent flame data, and coal-boiler NOx datasets; no full-scale, three-dimensional, high-Reynolds aerospace combustor case is demonstrated, so the aerospace pathway should be read as a research direction rather
  • A natural extension the review leaves implicit is to turn its acknowledged extrapolation error into an operational safeguard: equip PINN-based real-time controllers with Bayesian or ensemble uncertainty estimates that raise alarms when operating points leave the training distribution.
  • Combining CRK-PINN conservation constraints with quasi-steady-state stiffness reduction could make online reduced-order kinetics feasible for ammonia and hydrogen blends, where stiffness and non-equilibrium transport are the limiting factors.
  • The headline speedups apply to chemical source-term integration and to reactive-flow simulation components, not to the full engine design cycle; end-to-end savings would be smaller once training time, data acquisition, and validation costs are included.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. This review surveys physics-informed neural networks (PINNs) for clean combustion, with an emphasis on aerospace propulsion. It covers PINN fundamentals and loss formulations (Eqs. 1-4), applications to flame dynamics, turbulent combustion, emissions prediction, thermoacoustic instabilities, stiff chemistry, and control/optimization, and it compares PINNs with traditional CFD. The paper compiles recent references and reproduces selected quantitative claims from the literature, including CRK-PINN speedups (6.0-14.6x for source terms and 2.3-4.9x overall, Section 4.1), NOx prediction accuracy (R2 about 0.96, Table 2), and hybrid CFD-PINN concepts. It concludes that PINNs have emerged as a transformative framework for combustion modeling and asserts that next-generation aerospace engines will rely on PINNs.

Significance. The paper is a broad, readable survey that could serve as an entry point for researchers interested in PINN applications in combustion. Its strength is the breadth of coverage and the explicit discussion of challenges in Section 6. The central hedged claim that 'PINNs show potential' is consistent with the cited literature. However, the stronger framing in the abstract and conclusion significantly outruns the evidence: the surveyed applications are mainly canonical configurations (0-D/1-D/2-D/3-D flames, lab-scale combustors, coal-fired boilers) rather than aerospace propulsion systems. The paper also lacks a systematic review methodology, and the quantitative speedup/accuracy numbers are presented without critical appraisal of their baselines. These issues, together with a large duplicated section, currently limit the review's utility as a reliable synthesis.

major comments (3)
  1. [Abstract; §9 Conclusion] The abstract states that 'Next-generation aerospace engines rely on PINNs' and the conclusion calls PINNs a 'transformative framework for overcoming the limitations of traditional combustion modeling.' These categorical statements are not supported by the cited evidence. The applications in §4 are 0-D autoignition, 2-D Bunsen flames, 3-D turbulent jet flames, bluff-body combustors, and coal boilers, not engine-scale, high-Reynolds, multi-species propulsion cases. Moreover, §6.2 concedes in Eq. (53) that predictive errors increase upon extrapolation and in Eq. (54) that computational cost grows as exp(αN_params). Please replace the categorical wording with a hedged 'potential' claim and add an explicit discussion of the evidence gap between canonical flame studies and full aerospace propulsion.
  2. [§2.3, §4.1, Table 2] The quantitative performance claims are presented as if they demonstrate engine-level benefit, but they are scoped to specific surrogate sub-problems. The 'nearly ten times faster' claim in §2.3 is attributed to Jeon et al. [19], a hybrid FVM/network method rather than a PINN applied to an aerospace geometry. In §4.1, the '6.0–14.6 times faster' figure from [11] concerns chemical source term computation, and '2.3–4.9 times overall simulation speedups' is relative to direct integration on canonical 0-D/2-D/3-D flame cases; neither is a full CFD or engine benchmark. Table 2's R²≈0.96 is from coal-boiler NOx prediction, not aerospace propulsion. Each of these numbers should be accompanied by its precise scope, and the paper should state that no coupled LES or engine-scale benchmark is available in the reviewed literature.
  3. [§4.6, §5.5–5.6] The real-time control and hybrid-LES claims are largely aspirational. Equations (20)–(25) formulate PINN-based model predictive control but no cited study is shown to implement a closed-loop combustion controller with a PINN. Similarly, Eqs. (34)–(35) and (47) present robust control and subgrid-scale closure frameworks without a demonstrated application. The text should clearly separate demonstrated results from proposed research directions, especially because the abstract's claim about 'enabling real-time control methods' rests on these unvalidated formulations.
minor comments (5)
  1. [§7 and §8] Sections 7 and 8 are near-verbatim duplicates (e.g., §7.1 vs §8.1, §7.2 vs §8.2, §7.3 vs §8.3, etc., with repeated equations). This is a serious editing error that must be fixed by consolidating into one future-perspectives section.
  2. [Abstract] The abstract uses 'Physically Informed Neural Networks' while the rest of the paper uses 'Physics-Informed Neural Networks.' Please make the terminology consistent.
  3. [Eq. (12), Fig. 7] Equation (12) has incomplete square-root formatting, and the caption of Fig. 7 contains the typo 'PIINs' instead of 'PINNs.' Figure 7 also labels the methods inconsistently.
  4. [Table 2] The table reports R² and RMSE ranges but does not indicate the original sources per row, the data set size, or the operating conditions. Please add citations and clarify units; also fix the 'T raining' spacing in the header.
  5. [References] Several strong claims are supported by preprints or workshop papers (e.g., [97], [101]). Where peer-reviewed versions exist, please cite those instead.

Circularity Check

0 steps flagged

No significant circularity: this is a review importing independent external results; its own §6.2 limitation statement concedes extrapolation limits, and the authors' self-citations are not load-bearing.

full rationale

This paper is a review/survey with no original derivations, no fitted parameters, and no quantities that are fit and then re-labeled as predictions. The load-bearing speedup and accuracy numbers are taken from independent external studies: the 6.0–14.6x chemical-source-term speedups and 2.3–4.9x simulation speedups (Section 4.1, Section 4.5.4) are attributed to Zhang et al. [11], and the test R2≈0.96 NOx results (Table 2) are attributed to Zhu et al. [76] and Li et al. [74]. These are imported results from other groups, not outputs of the present review, so there is no fitted-input-called-prediction pattern. The PINN formulation in Section 2 (Eqs. 1–4) follows the standard Raissi et al. [5] and Karniadakis et al. [6] framework, which is foundational external work rather than a self-citation chain. The authors' own prior publications (e.g., refs. [40]–[42], [49], [50], [67]) appear in background discussions of ammonia/MILD combustion, fractal burners, and radiative heat loss; none is used to justify the central PINN claims, to import a uniqueness theorem, or to smuggle in an ansatz, so they do not constitute load-bearing self-citation. The paper's own Section 6.2 explicitly concedes that 'Predictive errors tend to increase when extrapolating beyond the training data distribution' (Eq. 53) and gives exponential cost growth in parameters (Eq. 54). That is an honest limitation statement, not a circularity; it indicates that the aerospace-propulsion extrapolation is a risk identified by the paper itself rather than a claim guaranteed by its construction. No equation-level circularity can be exhibited because there is no derivation chain: the review reports existing results and proposes future directions. Under the rule that self-citation becomes circularity only when the load-bearing argument reduces to an unverified self-citation, the self-citations here are not load-bearing. The most serious concern is overclaiming from canonical combustion cases to full aerospace regimes, but that is an extrapolation/correctness concern, not a circularity concern. Therefore the honest finding is no significant circularity, score 0.

Axiom & Free-Parameter Ledger

0 free parameters · 4 axioms · 0 invented entities

The paper introduces no fitted parameters and no invented entities; as a review, its only 'axioms' are the standard PINN loss formulation (Eq. 4), the governing conservation equations, the faithfulness of its citation summaries, and the extrapolation premise that laboratory-scale demonstrations carry over to aerospace propulsion. The last two are the ones a critical reader should interrogate.

axioms (4)
  • standard math The composite PINN loss function (Eq. 4: L = Ldata + λPDE LPDE + λBC LBC + λIC LIC) is a valid way to solve the governing PDEs
    Invoked throughout Section 2.1 as the foundation of all reviewed methods; this is the standard PINN framework from the cited literature.
  • standard math Conservation equations for mass, momentum, energy, and species (e.g., Eq. 38 and Eq. 9) correctly describe clean combustion phenomena
    Sections 3 and 4 assume these PDEs are the right physics for flames, emissions, and thermoacoustics; the review never questions the underlying combustion models.
  • domain assumption The quantitative results cited from the literature (speedups, R2 values, field reconstructions) are accurately and faithfully summarized
    Every concrete claim in Sections 4 and 5 rests on this, and the paper provides no independent verification of Zhang et al. [11], Mariappan et al. [71], or Zhu et al. [76].
  • domain assumption Results on canonical flames and laboratory combustors extrapolate to aerospace propulsion conditions
    This is the 'pathway to sustainable aerospace propulsion' premise. It is contradicted in spirit by the paper's own Eq. 53 (generalization error grows outside the training distribution) and Eq. 54 (exponential cost scaling), and it is never tested on a propulsion-scale case.

pith-pipeline@v1.3.0-alltime-deepseek · 23560 in / 16712 out tokens · 169095 ms · 2026-08-04T21:17:15.994797+00:00 · methodology

0 comments
Cite this review

Pith. "Pith review of Physics-Informed Neural Networks in Clean Combustion: A Pathway to Sustainable Aerospace Propulsion." pith.science (2026). https://pith.science/paper/QH3W2H7R

@misc{pith2026250908094,
  author       = {Pith},
  title        = {Pith review of: Physics-Informed Neural Networks in Clean Combustion: A Pathway to Sustainable Aerospace Propulsion},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/QH3W2H7R}},
  note         = {Machine review of arXiv:2509.08094}
}
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Achieving clean combustion systems is crucial in terms of solving environmental impacts, decarbonization needs and sustainability matters. Traditional combustion modeling techniques via computational fluid dynamics with accurate chemical kinetics face obstacles in computational cost and accurate representation of turbulence-chemistry interactions. Physically Informed Neural Networks (PINNs) as a new framework, merges physical laws with data-driven learning and shows great potential as an alternative methodology. By directly integrating conservation equations into their training process, PINNs achieve accurate mesh-free modeling of complex combustion phenomena despite having limited data sets. This review examines how this approach applies to clean combustion systems while focusing on their impact in aerospace applications including flame dynamics, turbulent combustion, emission prediction, and instability management in propulsion systems. Next-generation aerospace engines rely on PINNs to reduce computational costs while increasing predictive performance and enabling real-time control methods. This analysis concludes by exploring current barriers and future paths, while demonstrating how PINNs can revolutionize sustainable and efficient combustion technologies in aerospace propulsion systems.

Figures

Figures reproduced from arXiv: 2509.08094 by Bok Jik Lee, Caleb Caldwell, Jacob Baltes, Mahmood Mousavi, Muteb Aljasem.

Figure 1
Figure 1. Figure 1: Schematic diagram of a PINNs. The neural network takes spatial and temporal [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Algorithmic framework of PINNs. Adapted from Liu et al. [ [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Schematic of PINNs for inverse problems in thermoacoustic systems. The diagram [PITH_FULL_IMAGE:figures/full_fig_p011_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: PINNs solution from experimental data showing (a-d) comparison of acoustic pres [PITH_FULL_IMAGE:figures/full_fig_p013_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Probability density functions of the LOM parameters and train errors predicted [PITH_FULL_IMAGE:figures/full_fig_p014_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Time-continuous predictions of the autoignition process via CRK-PINNs and data [PITH_FULL_IMAGE:figures/full_fig_p016_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Comparison of DPINNs and PINNs solutions against the reference at four time levels. [PITH_FULL_IMAGE:figures/full_fig_p016_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Comparison of loss functions for different PINN-based methods, including DPINN. [PITH_FULL_IMAGE:figures/full_fig_p017_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: PINN-based NOx emission prediction framework from Zhu et al. [76] [PITH_FULL_IMAGE:figures/full_fig_p021_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Schematic of the two PINN-DeepONet networks as presented by Taassob et al. [ [PITH_FULL_IMAGE:figures/full_fig_p021_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Comparisons of density contours for Flames 57, 59, 80, and 103 [ [PITH_FULL_IMAGE:figures/full_fig_p022_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: Comparisons of radial velocity contours for Flames 57, 59, and 80 [ [PITH_FULL_IMAGE:figures/full_fig_p022_12.png] view at source ↗
Figure 13
Figure 13. Figure 13: Schematic of the Quantum Physics-Informed Neural Network (QPINN) hybrid [PITH_FULL_IMAGE:figures/full_fig_p032_13.png] view at source ↗
Figure 14
Figure 14. Figure 14: Continuous Variable (CV) QPINNs framework for solving PDEs using Gaussian and [PITH_FULL_IMAGE:figures/full_fig_p033_14.png] view at source ↗

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

108 extracted references · 76 canonical work pages · 10 internal anchors

  1. [1]

    Introduction to combustion, volume 287

    Stephen R Turns et al. Introduction to combustion, volume 287. McGraw-Hill Companies New York, NY, USA, 1996

    1996

  2. [2]

    Combustion

    Irvin Glassman, Richard A Yetter, and Nick G Glumac. Combustion. Academic press, 2014

    2014

  3. [3]

    Theoretical and numerical combustion

    Thierry Poinsot. Theoretical and numerical combustion. RT Edwards, 2005

    2005

  4. [4]

    Combustion

    Jurgen Warnatz, Ulrich Maas, Robert W Dibble, and J Warnatz. Combustion. Springer, 2006

    2006

  5. [5]

    Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations

    Maziar Raissi, Paris Perdikaris, and George E Karniadakis. 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. 34

    2019

  6. [6]

    Physics-informed machine learning

    George Em Karniadakis, Ioannis G Kevrekidis, Lu Lu, Paris Perdikaris, Sifan Wang, and Liu Yang. Physics-informed machine learning. Nature Reviews Physics , 3(6):422–440, 2021

    2021

  7. [7]

    Scientific machine learning through physics–informed neu- ral networks: Where we are and what’s next

    Salvatore Cuomo, Vincenzo Schiano Di Cola, Fabio Giampaolo, Gianluigi Rozza, Maziar Raissi, and Francesco Piccialli. Scientific machine learning through physics–informed neu- ral networks: Where we are and what’s next. Journal of Scientific Computing , 92(3):88, 2022

    2022

  8. [8]

    A physics-informed deep convolutional neural network for simulating and predicting transient darcy flows in heterogeneous reservoirs without labeled data

    Zhao Zhang. A physics-informed deep convolutional neural network for simulating and predicting transient darcy flows in heterogeneous reservoirs without labeled data. Journal of Petroleum Science and Engineering , 211:110179, 2022

    2022

  9. [9]

    Physics informed integral neural network for dynamic modelling of solvent-based post-combustion co2 capture process

    Peng Sha, Cheng Zheng, Xiao Wu, and Jiong Shen. Physics informed integral neural network for dynamic modelling of solvent-based post-combustion co2 capture process. Applied Energy, 377:124344, 2025

    2025

  10. [10]

    Model predictive control of diesel engine emissions based on neural network modeling

    Jiadi Zhang, Xiao Li, Ilya Kolmanovsky, Munechika Tsutsumi, and Hayato Nakada. Model predictive control of diesel engine emissions based on neural network modeling. In 2024 American Control Conference (ACC) , pages 4268–4274. IEEE, 2024

    2024

  11. [11]

    Crk-pinn: A physics-informed neural net- work for solving combustion reaction kinetics ordinary differential equations

    Shihong Zhang, Chi Zhang, and Bosen Wang. Crk-pinn: A physics-informed neural net- work for solving combustion reaction kinetics ordinary differential equations. Combustion and Flame , 269:113647, 2024

    2024

  12. [12]

    Exploring surface reaction mech- anism using a surface reaction neural network framework

    Lin Luo, Qimin Liu, Junhao Sun, and Yaosong Huang. Exploring surface reaction mech- anism using a surface reaction neural network framework. Chemical Engineering Science , page 121307, 2025

    2025

  13. [13]

    Ppinn: Parareal physics-informed neural network for time-dependent pdes

    Xuhui Meng, Zhen Li, Dongkun Zhang, and George Em Karniadakis. Ppinn: Parareal physics-informed neural network for time-dependent pdes. Computer Methods in Applied Mechanics and Engineering , 370:113250, 2020

    2020

  14. [14]

    An adaptive sampling method based on expected improvement function and residual gradient in pinns

    Yanbing Liu, Liping Chen, Jianwan Ding, and Yu Chen. An adaptive sampling method based on expected improvement function and residual gradient in pinns. IEEE Access, 2024

    2024

  15. [15]

    Physics-informed neural net- works for turbulent combustion: Toward extracting more statistics and closure from point multiscalar measurements

    Arsalan Taassob, Rishikesh Ranade, and Tarek Echekki. Physics-informed neural net- works for turbulent combustion: Toward extracting more statistics and closure from point multiscalar measurements. Energy & Fuels , 37(22):17484–17498, 2023

    2023

  16. [16]

    Physics-informed neural network for solving a one-dimensional solid mechanics problem

    Vishal Singh, Dineshkumar Harursampath, Sharanjeet Dhawan, Manoj Sahni, Sahaj Sax- ena, and Rajnish Mallick. Physics-informed neural network for solving a one-dimensional solid mechanics problem. Modelling, 5(4):1532–1549, 2024

    2024

  17. [17]

    Physics-informed neural network solver for numerical analysis in geoengineering

    Xiao-Xuan Chen, Pin Zhang, and Zhen-Yu Yin. Physics-informed neural network solver for numerical analysis in geoengineering. Georisk: Assessment and Management of Risk for Engineered Systems and Geohazards , 18(1):33–51, 2024

    2024

  18. [18]

    Understanding physics-informed neural networks: techniques, applications, trends, and challenges

    Amer Farea, Olli Yli-Harja, and Frank Emmert-Streib. Understanding physics-informed neural networks: techniques, applications, trends, and challenges. AI, 5(3):1534–1557, 2024

    2024

  19. [19]

    Finite volume method network for the acceleration of unsteady computational fluid dynamics: Non-reacting and reacting flows

    Joongoo Jeon, Juhyeong Lee, and Sung Joong Kim. Finite volume method network for the acceleration of unsteady computational fluid dynamics: Non-reacting and reacting flows. International Journal of Energy Research , 46(8):10770–10795, 2022. 35

    2022

  20. [20]

    Flamepinn-1d: Physics-informed neural networks to solve forward and inverse problems of 1d laminar flames

    Jiahao Wu, Su Zhang, Yuxin Wu, Guihua Zhang, Xin Li, and Hai Zhang. Flamepinn-1d: Physics-informed neural networks to solve forward and inverse problems of 1d laminar flames. Combustion and Flame , 273:113964, 2025

    2025

  21. [21]

    Exploring Physics-Informed Neural Networks: From Fundamentals to Applications in Complex Systems

    Sai Ganga and Ziya Uddin. Exploring physics-informed neural networks: From funda- mentals to applications in complex systems. arXiv preprint arXiv:2410.00422 , 2024

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  22. [22]

    On the Generalization of PINNs outside the training domain and the Hyperparameters influencing it

    A Bonfanti, R Santana, M Ellero, and B Gholami. On the generalization of pinns out- side the training domain and the hyperparameters influencing it,(2023). arXiv preprint arXiv:2302.07557

    work page internal anchor Pith review Pith/arXiv arXiv 2023

  23. [23]

    Exploring physics-informed neural networks for the generalized nonlinear sine-gordon equation

    Alemayehu Tamirie Deresse and Tamirat Temesgen Dufera. Exploring physics-informed neural networks for the generalized nonlinear sine-gordon equation. Applied Computational Intelligence and Soft Computing , 2024(1):3328977, 2024

    2024

  24. [24]

    Generalization of PINNs for various boundary and initial conditions

    V Schäfer, M Bracke, R Pinnau, and M Siggel. Generalization of PINNs for various boundary and initial conditions . PhD thesis, Master Thesis, University of Kaiserslautern, 2022

    2022

  25. [25]

    Estimates on the generalization error of physics-informed neural networks for approximating pdes

    Siddhartha Mishra and Roberto Molinaro. Estimates on the generalization error of physics-informed neural networks for approximating pdes. IMA Journal of Numerical Analysis, 43(1):1–43, 2023

    2023

  26. [26]

    Physics- informed neural networks for inverse problems in supersonic flows

    Ameya D Jagtap, Zhiping Mao, Nikolaus Adams, and George Em Karniadakis. Physics- informed neural networks for inverse problems in supersonic flows. Journal of Computa- tional Physics , 466:111402, 2022

    2022

  27. [27]

    The Challenges of the Nonlinear Regime for Physics-Informed Neural Networks

    Andrea Bonfanti, Giuseppe Bruno, and Cristina Cipriani. The challenges of the nonlinear regime for physics-informed neural networks. arXiv preprint arXiv:2402.03864 , 2024

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  28. [28]

    Investigating and Mitigating Failure Modes in Physics-informed Neural Networks (PINNs)

    Shamsulhaq Basir. Investigating and mitigating failure modes in physics-informed neural networks (pinns). arXiv preprint arXiv:2209.09988 , 2022

    work page internal anchor Pith review Pith/arXiv arXiv 2022

  29. [29]

    Approximation of Large Stiff Acausal Models

    Ranjan Anantharaman. Approximation of Large Stiff Acausal Models . Massachusetts Institute of Technology, 2023

    2023

  30. [30]

    Tackling the curse of dimensionality with physics-informed neural networks

    Zheyuan Hu, Khemraj Shukla, George Em Karniadakis, and Kenji Kawaguchi. Tackling the curse of dimensionality with physics-informed neural networks. Neural Networks , 176:106369, 2024

    2024

  31. [31]

    On the importance of the mathematical formulation to get pinns working

    Brahim El Mokhtari, Cédric Chauviere, and Pierre Bonnet. On the importance of the mathematical formulation to get pinns working. IEEE Transactions on Electromagnetic Compatibility, 2024

    2024

  32. [32]

    Locally adaptive ac- tivation functions with slope recovery for deep and physics-informed neural networks

    Ameya D Jagtap, Kenji Kawaguchi, and George Em Karniadakis. Locally adaptive ac- tivation functions with slope recovery for deep and physics-informed neural networks. Proceedings of the Royal Society A , 476(2239):20200334, 2020

    2020

  33. [33]

    Auto-pinn: understanding and optimizing physics-informed neural architecture

    Yicheng Wang, Xiaotian Han, Chia-Yuan Chang, Daochen Zha, Ulisses Braga-Neto, and Xia Hu. Auto-pinn: understanding and optimizing physics-informed neural architecture. arXiv preprint arXiv:2205.13748 , 2022

    Pith/arXiv arXiv 2022

  34. [34]

    Separable physics-informed neural networks

    Junwoo Cho, Seungtae Nam, Hyunmo Yang, Seok-Bae Yun, Youngjoon Hong, and Eun- byung Park. Separable physics-informed neural networks. Advances in Neural Information Processing Systems, 36, 2024. 36

    work page 2024

  35. [35]

    Robust regres- sion with highly corrupted data via physics informed neural networks

    Wei Peng, Wen Yao, Weien Zhou, Xiaoya Zhang, and Weijie Yao. Robust regres- sion with highly corrupted data via physics informed neural networks. arXiv preprint arXiv:2210.10646, 2022

    Pith/arXiv arXiv 2022

  36. [36]

    Physics-informed neural networks with unknown measurement noise

    Philipp Pilar and Niklas Wahlström. Physics-informed neural networks with unknown measurement noise. In 6th Annual Learning for Dynamics & Control Conference , pages 235–247. PMLR, 2024

    work page 2024

  37. [37]

    On the Preprocessing of Physics-informed Neural Networks: How to Better Utilize Data in Fluid Mechanics

    Shengfeng Xu, Chang Yan, Zhenxu Sun, Renfang Huang, Dilong Guo, and Guowei Yang. On the preprocessing of physics-informed neural networks: How to better utilize data in fluid mechanics. arXiv preprint arXiv:2403.19923 , 2024

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  38. [38]

    Cleaner combustion

    Frédérique Battin-Leclerc, John M Simmie, and Edward Blurock. Cleaner combustion. Developing Detailed Chemical Kinetic Models. Series: Green Energy and Technology. Cham: Springer International Publishing AG , 2013

    work page 2013

  39. [39]

    What fuel properties enable higher thermal efficiency in spark- ignited engines? Progress in Energy and Combustion Science , 82:100876, 2021

    James P Szybist, Stephen Busch, Robert L McCormick, Josh A Pihl, Derek A Splitter, Matthew A Ratcliff, Christopher P Kolodziej, John ME Storey, Melanie Moses-DeBusk, David Vuilleumier, et al. What fuel properties enable higher thermal efficiency in spark- ignited engines? Progress in Energy and Combustion Science , 82:100876, 2021

    work page 2021

  40. [40]

    On the effects of adding syngas to an ammonia-mild combustion regime—a computational study of the reaction zone structure

    S Mahmood Mousavi, Bok Jik Lee, Jinyoung Kim, Freshteh Sotoudeh, Byoungjoo Chun, Daeyoung Jun, Nader Karimi, and Javad Abolfazli Esfahani. On the effects of adding syngas to an ammonia-mild combustion regime—a computational study of the reaction zone structure. International Journal of Hydrogen Energy , 52:226–240, 2024

    work page 2024

  41. [41]

    A comprehensive investigation of acoustic power level in a moderate or intense low oxygen dilution in a jet-in-hot-coflow under various working conditions

    Seyed Mahmood Mousavi, Reza Kamali, Freshteh Sotoudeh, Reza Pourabidi, Nader Karimi, and In-Seuck Jeung. A comprehensive investigation of acoustic power level in a moderate or intense low oxygen dilution in a jet-in-hot-coflow under various working conditions. Aerospace science and technology , 93:105339, 2019

    work page 2019

  42. [42]

    On the effects of nh3 addition to a reacting mixture of h2/ch4 under mild combustion regime: Numerical modeling with a modified edc combus- tion model

    Seyed Mahmood Mousavi, Freshteh Sotoudeh, Daeyoung Jun, Bok Jik Lee, Javad Abol- fazli Esfahani, and Nader Karimi. On the effects of nh3 addition to a reacting mixture of h2/ch4 under mild combustion regime: Numerical modeling with a modified edc combus- tion model. Fuel, 326:125096, 2022

    work page 2022

  43. [43]

    Sustainable energy transition for renewable and low carbon grid electricity generation and supply

    Moses Jeremiah Barasa Kabeyi and Oludolapo Akanni Olanrewaju. Sustainable energy transition for renewable and low carbon grid electricity generation and supply. Frontiers in Energy research , 9:743114, 2022

    work page 2022

  44. [44]

    Fuels of the Future for Renewable Energy Sources (Ammonia, Biofuels, Hydrogen)

    G Ali Mansoori, L Barnie Agyarko, L Antonio Estevez, Behrooz Fallahi, Georgi Glady- shev, Ronaldo Gonçalves dos Santos, Shawn Niaki, Ognjen Perišić, Mika Sillanpää, Kaniki Tumba, et al. Fuels of the future for renewable energy sources (ammonia, biofuels, hy- drogen). arXiv preprint arXiv:2102.00439 , 2021

    work page internal anchor Pith review Pith/arXiv arXiv 2021

  45. [45]

    A review on alternative fuels in future energy system

    Hrvoje Stančin, Hrvoje Mikulčić, Xuebin Wang, and Neven Duić. A review on alternative fuels in future energy system. Renewable and sustainable energy reviews , 128:109927, 2020

    work page 2020

  46. [46]

    Impact of hydrogen on the environment

    Janusz Nowotny and T Nejat Veziroglu. Impact of hydrogen on the environment. Inter- national Journal of Hydrogen Energy , 36(20):13218–13224, 2011

    work page 2011

  47. [47]

    Hydrogen combustion, production, and applications: A review

    Mohamed A Habib, Gubran AQ Abdulrahman, Awad BS Alquaity, and Naef AA Qasem. Hydrogen combustion, production, and applications: A review. Alexandria Engineering Journal, 100:182–207, 2024. 37

    work page 2024

  48. [48]

    Low nox-lpg staged combustion double swirl flames

    Ayman M Elbaz, HA Moneib, KM Shebil, and William L Roberts. Low nox-lpg staged combustion double swirl flames. Renewable Energy, 138:303–315, 2019

    work page 2019

  49. [49]

    On the effects of fractal geometry on reacting and nonreacting flows in a low- swirl burner: A numerical study with large-eddy simulation

    Byoungjoo Chun, S Mahmood Mousavi, Jongkwon Lee, Bok Jik Lee, and Salah A Faroughi. On the effects of fractal geometry on reacting and nonreacting flows in a low- swirl burner: A numerical study with large-eddy simulation. Case Studies in Thermal Engineering, 49:103385, 2023

    work page 2023

  50. [50]

    Moderate or intense low-oxygen dilution combustion of methane diluted by co2 and n2

    Jianpeng Zhang, Jianchun Mi, Pengfei Li, Feifei Wang, and Bassam B Dally. Moderate or intense low-oxygen dilution combustion of methane diluted by co2 and n2. Energy & Fuels, 29(7):4576–4585, 2015

    work page 2015

  51. [51]

    Ammonia combustion and emissions in practical applications: a review

    Mohammad Alnajideen, Hao Shi, William Northrop, David Emberson, Seamus Kane, Pawel Czyzewski, Mustafa Alnaeli, Syed Mashruk, Kevin Rouwenhorst, Chunkan Yu, et al. Ammonia combustion and emissions in practical applications: a review. Carbon Neutrality, 3(1):1–45, 2024

    work page 2024

  52. [52]

    Ammonia combustion in furnaces: A review

    A Valera-Medina, MO Vigueras-Zuniga, H Shi, S Mashruk, M Alnajideen, A Alnasif, J Davies, Y Wang, X Zhu, W Yang, et al. Ammonia combustion in furnaces: A review. International Journal of Hydrogen Energy , 49:1597–1618, 2024

    work page 2024

  53. [53]

    Science and technology of ammonia combustion

    Hideaki Kobayashi, Akihiro Hayakawa, KD Kunkuma A Somarathne, and Ekenechukwu C Okafor. Science and technology of ammonia combustion. Proceedings of the combustion institute, 37(1):109–133, 2019

    work page 2019

  54. [54]

    A review on ammonia blends combustion for industrial applications

    Lianwei Kang, Weiguo Pan, Jiakai Zhang, Wenhuan Wang, and Congwei Tang. A review on ammonia blends combustion for industrial applications. Fuel, 332:126150, 2023

    work page 2023

  55. [55]

    Green synthetic fuels: Renewable routes for the conversion of non-fossil feedstocks into gaseous fuels and their end uses

    Elena Rozzi, Francesco Demetrio Minuto, Andrea Lanzini, and Pierluigi Leone. Green synthetic fuels: Renewable routes for the conversion of non-fossil feedstocks into gaseous fuels and their end uses. Energies, 13(2):420, 2020

    work page 2020

  56. [56]

    Biofuels an alternative to traditional fossil fuels: A comprehensive review

    Lubhan Cherwoo, Ishika Gupta, G Flora, Ritu Verma, Muskaan Kapil, Shailendra Kumar Arya, Balasubramani Ravindran, Kuan Shiong Khoo, Shashi Kant Bhatia, Soon Woong Chang, et al. Biofuels an alternative to traditional fossil fuels: A comprehensive review. Sustainable Energy Technologies and Assessments , 60:103503, 2023

    work page 2023

  57. [57]

    Biofuels: An alternative to conventional fuel and energy source

    Prayagraj Singh Deora, Yash Verma, Ram A vtar Muhal, Chandramani Goswami, Tej Singh, et al. Biofuels: An alternative to conventional fuel and energy source. Materials Today: Proceedings, 48:1178–1184, 2022

    work page 2022

  58. [58]

    The feasibility of synthetic fuels in renewable energy systems

    Iva Ridjan, Brian Vad Mathiesen, David Connolly, and Neven Duić. The feasibility of synthetic fuels in renewable energy systems. Energy, 57:76–84, 2013

    work page 2013

  59. [59]

    Fossil fuels and alternative fuels

    İbrahim Dinçer, Calin Zamfirescu, İbrahim Dinçer, and Calin Zamfirescu. Fossil fuels and alternative fuels. Sustainable Energy Systems and Applications , pages 169–201, 2012

    work page 2012

  60. [60]

    Oxy-fuel combustion technology: current status, applications, and trends

    Medhat A Nemitallah, Mohamed A Habib, Hassan M Badr, Syed A Said, Aqil Jamal, Rached Ben-Mansour, Esmail MA Mokheimer, and K Mezghani. Oxy-fuel combustion technology: current status, applications, and trends. International Journal of Energy Research, 41(12):1670–1708, 2017

    work page 2017

  61. [61]

    Oxy-fuel combustion of solid fuels

    Maja B Toftegaard, Jacob Brix, Peter A Jensen, Peter Glarborg, and Anker D Jensen. Oxy-fuel combustion of solid fuels. Progress in energy and combustion science , 36(5):581– 625, 2010. 38

    work page 2010

  62. [62]

    Oxy-fuel coal combustion—a review of the current state-of-the-art

    Günter Scheffknecht, Leema Al-Makhadmeh, Uwe Schnell, and Jörg Maier. Oxy-fuel coal combustion—a review of the current state-of-the-art. International Journal of Greenhouse Gas Control , 5:S16–S35, 2011

    work page 2011

  63. [63]

    Oxyfuel combustion for clean energy applications

    Mohamed A Habib, Medhat A Nemitallah, and Hassan M Badr. Oxyfuel combustion for clean energy applications . Springer, 2019

    work page 2019

  64. [64]

    Technological, economic, and emission analysis of the oxy-combustion process

    Brenda Raho, Marcello Giangreco, Gianpiero Colangelo, Marco Milanese, and Arturo de Risi. Technological, economic, and emission analysis of the oxy-combustion process. Applied Energy, 378:124821, 2025

    work page 2025

  65. [65]

    Flame stretch effects on partially premixed flames

    Chun W Choi and Ishwar K Puri. Flame stretch effects on partially premixed flames. Combustion and flame , 123(1-2):119–139, 2000

    work page 2000

  66. [66]

    Flame dynamics

    Moshe Matalon. Flame dynamics. Proceedings of the Combustion Institute , 32(1):57–82, 2009

    work page 2009

  67. [67]

    Large eddy simulation of the effects of radiative heat loss on combustion instability prediction

    Jongkwon Lee, Daeyoung Jun, Byoungjoo Chun, S Mahmood Mousavi, Bok Jik Lee, and Salah A Faroughi. Large eddy simulation of the effects of radiative heat loss on combustion instability prediction. Acta Astronautica, 217:312–322, 2024

    work page 2024

  68. [69]

    Turbulent flows

    Stephen B Pope. Turbulent flows. Measurement Science and Technology , 12(11):2020– 2021, 2001

    work page 2020

  69. [70]

    A pinn-deeponet framework for extracting turbulent combustion closure from multiscalar measurements

    Arsalan Taassob, Anuj Kumar, Kevin M Gitushi, Rishikesh Ranade, and Tarek Echekki. A pinn-deeponet framework for extracting turbulent combustion closure from multiscalar measurements. Computer Methods in Applied Mechanics and Engineering , 429:117163, 2024

    work page 2024

  70. [71]

    Learning thermoa- coustic interactions in combustors using a physics-informed neural network

    Sathesh Mariappan, Kamaljyoti Nath, and George Em Karniadakis. Learning thermoa- coustic interactions in combustors using a physics-informed neural network. Engineering Applications of Artificial Intelligence , 138:109388, 2024

    work page 2024

  71. [72]

    Predicting bifurcation and amplitude death characteristics of thermoacoustic instabilities from pinns-derived van der pol oscillators

    Mingke Xie, Xinyu Zhao, Dan Zhao, Jianqin Fu, Cody Shelton, and Bernhard Semlitsch. Predicting bifurcation and amplitude death characteristics of thermoacoustic instabilities from pinns-derived van der pol oscillators. Journal of Fluid Mechanics , 998:A46, 2024

    work page 2024

  72. [73]

    Efficient nonlinear flame response modeling for propulsion thermoacoustic analysis using limited numerical data

    Jiawei Wu, Teng Wang, Jiaqi Nan, Lijun Yang, and Jingxuan Li. A dual-path neural network model to construct the flame nonlinear thermoacoustic response in the time domain. arXiv preprint arXiv:2409.05885 , 2024

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  73. [74]

    Physics-Informed Deep Learning to Reduce the Bias in Joint Prediction of Nitrogen Oxides

    Lianfa Li, Roxana Khalili, Frederick Lurmann, Nathan Pavlovic, Jun Wu, Yan Xu, Yisi Liu, Karl O’Sharkey, Beate Ritz, Luke Oman, et al. Physics-informed deep learning to reduce the bias in joint prediction of nitrogen oxides. arXiv preprint arXiv:2308.07441 , 2023

    work page internal anchor Pith review Pith/arXiv arXiv 2023

  74. [75]

    Physics-based deep learning reveals rising heating demand heightens air pollution in Norwegian cities

    Cong Cao, Ramit Debnath, and R Michael Alvarez. Physics-based deep learning re- veals rising heating demand heightens air pollution in norwegian cities. arXiv preprint arXiv:2405.04716, 2024. 39

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  75. [76]

    A physics-informed neural net- work that considers monotonic relationships for predicting nox emissions from coal-fired boilers

    Baoyu Zhu, Shaojun Ren, Qihang Weng, and Fengqi Si. A physics-informed neural net- work that considers monotonic relationships for predicting nox emissions from coal-fired boilers. Fuel, 364:131026, 2024

    work page 2024

  76. [77]

    Reconstructing soot fields in acoustically forced laminar sooting flames using physics-informed machine learn- ing models

    Shiyu Liu, Haiou Wang, Zhiwei Sun, Kae Ken Foo, Graham J Nathan, Xue Dong, Michael J Evans, Bassam B Dally, Kun Luo, and Jianren Fan. Reconstructing soot fields in acoustically forced laminar sooting flames using physics-informed machine learn- ing models. Proceedings of the Combustion Institute , 40(1-4):105314, 2024

    work page 2024

  77. [78]

    Soot temperature and volume fraction field predictions via line-of-sight soot integral radiation equation informed neural networks in laminar sooting flames

    Qianlong Wang, Mingxue Gong, Alexis Matynia, Linghui Zhang, Yingyu Qian, and Chao Dang. Soot temperature and volume fraction field predictions via line-of-sight soot integral radiation equation informed neural networks in laminar sooting flames. Physics of Fluids , 36(12), 2024

    work page 2024

  78. [79]

    Accelerating the Chemical Kinetics Calculations in Combustion Simulations Using Physics Informed Neural Networks

    Ahmed MZ Almeldein. Accelerating the Chemical Kinetics Calculations in Combustion Simulations Using Physics Informed Neural Networks . PhD thesis, University of Mas- sachusetts Lowell, 2024

    work page 2024

  79. [80]

    Co+ oh→ co 2+ h: The relative reaction rate of five co isotopologues

    Karen L Feilberg, Stig R Sellevåg, Claus J Nielsen, David WT Griffith, and Matthew S Johnson. Co+ oh→ co 2+ h: The relative reaction rate of five co isotopologues. Physical Chemistry Chemical Physics , 4(19):4687–4693, 2002

    work page 2002

  80. [81]

    Instability behaviors and suppression of the unsteady autoignited turbulent jet flame in hot coflow

    Guijun Liu, Yuxin Wu, Suhui Li, and Yang Zhang. Instability behaviors and suppression of the unsteady autoignited turbulent jet flame in hot coflow. Combustion Science and Technology, 196(17):4923–4941, 2024

    work page 2024

Showing first 80 references.