Reduced-order modeling via convolutional autoencoder for emulating combustion of hydrogen/methane fuel blends
Numerical simulations are essential in comprehending the processes of flow and combustion in aerospace and power-generation systems. Yet, substantial computational demands render these simulations prohibitively expensive for design and optimization, owing to the large amount of design parameters to...
Gespeichert in:
| Veröffentlicht in: | Combustion and flame Jg. 274; S. 113981 |
|---|---|
| Hauptverfasser: | , , , , |
| Format: | Journal Article |
| Sprache: | Englisch |
| Veröffentlicht: |
Elsevier Inc
01.04.2025
|
| Schlagworte: | |
| ISSN: | 0010-2180 |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| Abstract | Numerical simulations are essential in comprehending the processes of flow and combustion in aerospace and power-generation systems. Yet, substantial computational demands render these simulations prohibitively expensive for design and optimization, owing to the large amount of design parameters to be surveyed in a wide design space. This study presents parametric reduced order models (ROMs) that leverage deep neural network-based dimension reduction through a convolutional autoencoder (AE) to emulate spatial distributions of physical flowfields in combustion problems. The proposed AE-based ROMs integrate multiple advancements, including design of experiment, nonlinear dimension reduction, and regression methods like kriging and deep neural networks (DNN) to enable parametric predictability. For comparison, proper orthogonal decomposition (POD)-based ROMs are also developed. Two distinct test scenarios are outlined: steam-diluted methane/hydrogen-blending oxy-combustion from a triple-coaxial nozzle and hydrogen-enriched combustion in a practical aero-derivative combustor. Results suggest that ROMs with kriging show superior performance against those with DNN. In both test scenarios, AE-based ROMs exhibit better prediction accuracy in emulating spatial distributions for physical variables of interest than POD-based ROMs. This is mainly due to the effective capture of nonlinear features by AE through intricate network structure. Such nonlinearity effect is examined by introducing interpretable “AE modes”, which demonstrate multiscale characteristics, distinct from POD modes, enabling a more detailed representation of local flow features. Both AE- and POD-based ROMs achieve a dramatic acceleration in prediction process, 8–9 orders of magnitude faster than traditional combustion simulations. While POD-based ROMs slightly underperform concerning prediction accuracy, they remain competitive in terms of training and prediction efficiency with a limited number of reduced bases.
The novelty of this research is the proposal of innovative parametric reduced order models (ROMs) that leverage deep neural network-based dimension reduction through a convolutional autoencoder (AE), for emulation of spatial distributions of physical variables in hydrogen-blended combustion problems. The proposed AE-based ROMs integrate design of experiment, nonlinear dimension reduction, and regression methods such as kriging and deep neural networks (DNN). Notably, they excel in capturing nonlinear features, through the introduction of interpretable “AE modes”, as opposed to intrinsically linear characteristics of POD modes.
It is significant because the constructed data-driven ROMs are expected to serve as reliable surrogates of numerical simulations, substantially reducing processing times while enhancing efficiency. These ROMs achieve online prediction time 8–9 orders of magnitude shorter than conventional combustion simulations, making them applicable across diverse engineering systems for design and optimization. |
|---|---|
| AbstractList | Numerical simulations are essential in comprehending the processes of flow and combustion in aerospace and power-generation systems. Yet, substantial computational demands render these simulations prohibitively expensive for design and optimization, owing to the large amount of design parameters to be surveyed in a wide design space. This study presents parametric reduced order models (ROMs) that leverage deep neural network-based dimension reduction through a convolutional autoencoder (AE) to emulate spatial distributions of physical flowfields in combustion problems. The proposed AE-based ROMs integrate multiple advancements, including design of experiment, nonlinear dimension reduction, and regression methods like kriging and deep neural networks (DNN) to enable parametric predictability. For comparison, proper orthogonal decomposition (POD)-based ROMs are also developed. Two distinct test scenarios are outlined: steam-diluted methane/hydrogen-blending oxy-combustion from a triple-coaxial nozzle and hydrogen-enriched combustion in a practical aero-derivative combustor. Results suggest that ROMs with kriging show superior performance against those with DNN. In both test scenarios, AE-based ROMs exhibit better prediction accuracy in emulating spatial distributions for physical variables of interest than POD-based ROMs. This is mainly due to the effective capture of nonlinear features by AE through intricate network structure. Such nonlinearity effect is examined by introducing interpretable “AE modes”, which demonstrate multiscale characteristics, distinct from POD modes, enabling a more detailed representation of local flow features. Both AE- and POD-based ROMs achieve a dramatic acceleration in prediction process, 8–9 orders of magnitude faster than traditional combustion simulations. While POD-based ROMs slightly underperform concerning prediction accuracy, they remain competitive in terms of training and prediction efficiency with a limited number of reduced bases.
The novelty of this research is the proposal of innovative parametric reduced order models (ROMs) that leverage deep neural network-based dimension reduction through a convolutional autoencoder (AE), for emulation of spatial distributions of physical variables in hydrogen-blended combustion problems. The proposed AE-based ROMs integrate design of experiment, nonlinear dimension reduction, and regression methods such as kriging and deep neural networks (DNN). Notably, they excel in capturing nonlinear features, through the introduction of interpretable “AE modes”, as opposed to intrinsically linear characteristics of POD modes.
It is significant because the constructed data-driven ROMs are expected to serve as reliable surrogates of numerical simulations, substantially reducing processing times while enhancing efficiency. These ROMs achieve online prediction time 8–9 orders of magnitude shorter than conventional combustion simulations, making them applicable across diverse engineering systems for design and optimization. |
| ArticleNumber | 113981 |
| Author | Ni, Chenxu Ding, Siyu Lu, Qingzhou Wang, Xingjian Chu, Xu |
| Author_xml | – sequence: 1 givenname: Siyu orcidid: 0000-0002-5143-8059 surname: Ding fullname: Ding, Siyu – sequence: 2 givenname: Chenxu surname: Ni fullname: Ni, Chenxu – sequence: 3 givenname: Xu surname: Chu fullname: Chu, Xu – sequence: 4 givenname: Qingzhou orcidid: 0009-0009-9022-1846 surname: Lu fullname: Lu, Qingzhou – sequence: 5 givenname: Xingjian surname: Wang fullname: Wang, Xingjian email: xingjianwang@tsinghua.edu.cn |
| BookMark | eNqNkMtOwzAQRb0oEuXxDxb7lHGch8sKxFuqhIRgbTn2uLhybOQ4lfh7EpUFYsVqNveemTknZBFiQEIuGKwYsOZyt9Kx78YhW696XJVQ1ivG-FqwBVkCMChKJuCYnAzDDgDaivMl6V_RjBpNEZPBRPto0LuwpXunqI5hH_2YXQzKUzXmiEHHOWZjotiPXuU5-7N2itFo6ceXSXGL4bLH_KECUjuip53HYIYzcmSVH_D8Z56S94f7t9unYvPy-Hx7syk0L6tcKKh4CZwrELZuG6UtmgbXChoFjNUtF2J6gJkWWlHXomvWnUVeiVpDp8qm5Kfk-sDVKQ5DQiu1y2q-MCflvGQgZ2VyJ38rk7MyeVA2Ia7-ID6T61X6-l_57lDG6cm9wyQH7SZ5aFxCnaWJ7j-YbySiljc |
| CitedBy_id | crossref_primary_10_1016_j_ijhydene_2025_151405 crossref_primary_10_1016_j_proci_2025_105796 crossref_primary_10_3390_pr13041093 crossref_primary_10_1016_j_applthermaleng_2025_127969 crossref_primary_10_1016_j_jcp_2025_114300 |
| Cites_doi | 10.1115/1.4049346 10.1016/j.combustflame.2022.112286 10.1006/jcph.2002.7146 10.1016/j.proci.2020.06.303 10.2514/1.J059877 10.1016/j.combustflame.2023.112743 10.1615/AtomizSpr.2020034830 10.1016/j.cma.2022.114764 10.1016/j.paerosci.2017.11.003 10.1016/j.combustflame.2024.113366 10.1016/j.proci.2020.06.045 10.2514/1.J060574 10.1038/s41467-023-42213-6 10.1016/j.combustflame.2016.03.027 10.1016/j.cma.2020.113379 10.2514/1.J057803 10.1016/j.cja.2024.08.012 10.1016/j.compfluid.2018.07.021 10.1016/j.pecs.2023.101073 10.1038/s42254-021-00314-5 10.1146/annurev.fl.25.010193.002543 10.1016/j.jcp.2019.108973 10.1016/j.cma.2018.07.017 10.1109/TNNLS.2021.3084827 10.1007/s10915-021-01462-7 10.1016/0169-7439(87)80084-9 10.1016/j.combustflame.2023.112900 10.1063/5.0020721 10.1016/j.paerosci.2005.02.001 10.1080/01621459.2017.1409123 10.1126/science.1127647 10.1063/5.0012906 10.1016/j.combustflame.2022.112600 10.1016/j.proci.2022.07.040 10.1007/s13748-019-00203-0 10.1038/s42256-021-00345-8 10.1137/130932715 10.1063/5.0217845 10.1146/annurev-fluid-010719-060214 10.1016/j.jcp.2019.01.031 10.1016/j.apenergy.2019.114159 |
| ContentType | Journal Article |
| Copyright | 2025 The Combustion Institute |
| Copyright_xml | – notice: 2025 The Combustion Institute |
| DBID | AAYXX CITATION |
| DOI | 10.1016/j.combustflame.2025.113981 |
| DatabaseName | CrossRef |
| DatabaseTitle | CrossRef |
| DatabaseTitleList | |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Engineering Chemistry |
| ExternalDocumentID | 10_1016_j_combustflame_2025_113981 S0010218025000197 |
| GroupedDBID | --- --K --M -~X .~1 0R~ 1B1 1~. 1~5 29F 4.4 41~ 457 4G. 5GY 5VS 7-5 71M 8P~ 9JN AABNK AACTN AAEDT AAEDW AAHCO AAIKJ AAKOC AALRI AAOAW AAQFI AAQXK AARJD AAXKI AAXUO ABDEX ABDMP ABFNM ABJNI ABMAC ABNUV ABTAH ABWVN ABXDB ACDAQ ACGFS ACIWK ACNCT ACNNM ACRLP ACRPL ADBBV ADEWK ADEZE ADMUD ADNMO ADTZH AEBSH AECPX AEIPS AEKER AENEX AFFNX AFJKZ AFTJW AGHFR AGUBO AGYEJ AHHHB AHIDL AHJVU AHPOS AI. AIEXJ AIKHN AITUG AKRWK AKURH ALMA_UNASSIGNED_HOLDINGS AMRAJ ANKPU ASPBG AVWKF AXJTR AZFZN BELTK BJAXD BKOJK BLXMC CS3 DU5 EBS EFJIC EJD ENUVR EO8 EO9 EP2 EP3 FDB FEDTE FGOYB FIRID FNPLU FYGXN G-2 G-Q GBLVA HVGLF HZ~ H~9 IHE J1W JARJE JJJVA KOM LY6 M41 MO0 N9A O-L O9- OAUVE OZT P-8 P-9 P2P PC. Q38 R2- RIG RNS ROL RPZ SAC SDF SDG SES SEW SPC SPCBC SSG SSR SST SSZ T5K T9H TN5 VH1 WUQ XPP ZMT ZY4 ~02 ~G- 9DU AATTM AAYWO AAYXX ACLOT ACVFH ADCNI AEUPX AFPUW AGQPQ AIGII AIIUN AKBMS AKYEP APXCP CITATION EFKBS EFLBG ~HD |
| ID | FETCH-LOGICAL-c324t-a0432033a08f576acfed6e9a06a011573880001d7078558b69bfe3485c0ba2623 |
| ISICitedReferencesCount | 7 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001416272900001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 0010-2180 |
| IngestDate | Tue Nov 18 21:02:17 EST 2025 Sat Nov 29 08:09:40 EST 2025 Sat Mar 22 15:53:15 EDT 2025 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Keywords | Proper orthogonal decomposition Nonlinear feature Convolutional autoencoder Reduced-order model (ROM) Kriging |
| Language | English |
| LinkModel | OpenURL |
| MergedId | FETCHMERGED-LOGICAL-c324t-a0432033a08f576acfed6e9a06a011573880001d7078558b69bfe3485c0ba2623 |
| ORCID | 0000-0002-5143-8059 0009-0009-9022-1846 |
| ParticipantIDs | crossref_citationtrail_10_1016_j_combustflame_2025_113981 crossref_primary_10_1016_j_combustflame_2025_113981 elsevier_sciencedirect_doi_10_1016_j_combustflame_2025_113981 |
| PublicationCentury | 2000 |
| PublicationDate | April 2025 2025-04-00 |
| PublicationDateYYYYMMDD | 2025-04-01 |
| PublicationDate_xml | – month: 04 year: 2025 text: April 2025 |
| PublicationDecade | 2020 |
| PublicationTitle | Combustion and flame |
| PublicationYear | 2025 |
| Publisher | Elsevier Inc |
| Publisher_xml | – name: Elsevier Inc |
| References | Jin, Kim (bib0002) 2024; 262 DeMers, Cottrell (bib0024) 1992; 5 Wold, Esbensen, Geladi (bib0013) 1987; 2 Guo, Hesthaven (bib0015) 2018; 341 Swischuk, Mainini, Peherstorfer, Willcox (bib0016) 2019; 179 Milano, Koumoutsakos (bib0025) 2002; 182 Fukami, Taira (bib0028) 2023; 14 Fresca, Dede, Manzoni (bib0037) 2021; 87 Hinton, Salakhutdinov (bib0022) 2006; 313 Fukami, Nakamura, Fukagata (bib0038) 2020; 32 Ebi, Clemens (bib0003) 2016; 168 Lee, Carlberg (bib0033) 2020; 404 Gruber, Gunzburger, Ju, Wang (bib0029) 2022; 393 Dhillon, Verma (bib0035) 2019; 9 M. Frenklach, C.T. Bowman, G.P. Smith, W.C. Gardiner, World Wide Web location. Noble, Wu, Emerson, Sheppard, Lieuwen, Angello (bib0004) 2021; 143 Milan, Torelli, Lusch, Magnotti (bib0031) 2020; 30 Mak, Sung, Wang, Yeh, Chang, Joseph, Yang, Wu (bib0039) 2018; 113 Chang, Zhang, Wang, Yeh, Mak, Sung, Jeff Wu, Yang (bib0017) 2019; 57 Version 3.0, 1999. Han, Hossain, Wang, Li, Xu (bib0032) 2020; 259 Wei, Zhang, Wang, Huang (bib0007) 2023; 249 Wang, Hesthaven, Ray (bib0011) 2019; 384 Mitra (bib0041) 2021; 3 Karniadakis, Kevrekidis, Lu, Perdikaris, Wang, Yang (bib0042) 2021; 3 Ding, Wang, Lu, Wang (bib0027) 2024; 37 Yondo, Andrés, Valero (bib0043) 2018; 96 Brunton, Noack, Koumoutsakos (bib0010) 2020; 52 Berkooz, Holmes, Lumley (bib0012) 1993; 25 Chang, Wang, Zhang, Li, Mak, Wu, Yang (bib0020) 2021; 59 Zhang, Zhao (bib0030) 2021; 59 Zhang, Lu, Yang (bib0005) 2023; 255 Agostini (bib0023) 2020; 32 Xia, Han, Wei, Zhang, Wang, Huang, Hasse (bib0006) 2023; 39 Wang, Chang, Li, Yang, Su (bib0018) 2021; 38 Queipo, Haftka, Shyy, Goel, Vaidyanathan, Kevin Tucker (bib0008) 2005; 41 Xu, Duraisamy (bib0026) 2020; 372 Aversano, Ferrarotti, Parente (bib0019) 2021; 38 Su-ungkavatin, Tiruta-Barna, Hamelin (bib0001) 2023; 96 Li, Liu, Yang, Peng, Zhou (bib0036) 2022; 33 Ni, Ding, Li, Chu, Ren, Wang (bib0021) 2024; 36 Benner, Gugercin, Willcox (bib0009) 2015; 57 Dai, Zhou, Zhang, Cheng, Liu, Zhao, Wang, Gao (bib0034) 2023; 252 Perry, Henry de Frahan, Yellapantula (bib0014) 2022; 244 Guo (10.1016/j.combustflame.2025.113981_bib0015) 2018; 341 Agostini (10.1016/j.combustflame.2025.113981_bib0023) 2020; 32 DeMers (10.1016/j.combustflame.2025.113981_bib0024) 1992; 5 Jin (10.1016/j.combustflame.2025.113981_bib0002) 2024; 262 Brunton (10.1016/j.combustflame.2025.113981_bib0010) 2020; 52 Ni (10.1016/j.combustflame.2025.113981_bib0021) 2024; 36 Berkooz (10.1016/j.combustflame.2025.113981_bib0012) 1993; 25 Wang (10.1016/j.combustflame.2025.113981_bib0018) 2021; 38 Zhang (10.1016/j.combustflame.2025.113981_bib0030) 2021; 59 Noble (10.1016/j.combustflame.2025.113981_bib0004) 2021; 143 Wei (10.1016/j.combustflame.2025.113981_bib0007) 2023; 249 Benner (10.1016/j.combustflame.2025.113981_bib0009) 2015; 57 Li (10.1016/j.combustflame.2025.113981_bib0036) 2022; 33 10.1016/j.combustflame.2025.113981_bib0040 Milano (10.1016/j.combustflame.2025.113981_bib0025) 2002; 182 Fukami (10.1016/j.combustflame.2025.113981_bib0028) 2023; 14 Gruber (10.1016/j.combustflame.2025.113981_bib0029) 2022; 393 Fukami (10.1016/j.combustflame.2025.113981_bib0038) 2020; 32 Mitra (10.1016/j.combustflame.2025.113981_bib0041) 2021; 3 Chang (10.1016/j.combustflame.2025.113981_bib0017) 2019; 57 Perry (10.1016/j.combustflame.2025.113981_bib0014) 2022; 244 Chang (10.1016/j.combustflame.2025.113981_bib0020) 2021; 59 Swischuk (10.1016/j.combustflame.2025.113981_bib0016) 2019; 179 Dai (10.1016/j.combustflame.2025.113981_bib0034) 2023; 252 Queipo (10.1016/j.combustflame.2025.113981_bib0008) 2005; 41 Ebi (10.1016/j.combustflame.2025.113981_bib0003) 2016; 168 Lee (10.1016/j.combustflame.2025.113981_bib0033) 2020; 404 Ding (10.1016/j.combustflame.2025.113981_bib0027) 2024; 37 Xia (10.1016/j.combustflame.2025.113981_bib0006) 2023; 39 Hinton (10.1016/j.combustflame.2025.113981_bib0022) 2006; 313 Fresca (10.1016/j.combustflame.2025.113981_bib0037) 2021; 87 Yondo (10.1016/j.combustflame.2025.113981_bib0043) 2018; 96 Wold (10.1016/j.combustflame.2025.113981_bib0013) 1987; 2 Mak (10.1016/j.combustflame.2025.113981_bib0039) 2018; 113 Zhang (10.1016/j.combustflame.2025.113981_bib0005) 2023; 255 Xu (10.1016/j.combustflame.2025.113981_bib0026) 2020; 372 Milan (10.1016/j.combustflame.2025.113981_bib0031) 2020; 30 Han (10.1016/j.combustflame.2025.113981_bib0032) 2020; 259 Aversano (10.1016/j.combustflame.2025.113981_bib0019) 2021; 38 Karniadakis (10.1016/j.combustflame.2025.113981_bib0042) 2021; 3 Dhillon (10.1016/j.combustflame.2025.113981_bib0035) 2019; 9 Wang (10.1016/j.combustflame.2025.113981_bib0011) 2019; 384 Su-ungkavatin (10.1016/j.combustflame.2025.113981_bib0001) 2023; 96 |
| References_xml | – volume: 38 start-page: 5373 year: 2021 end-page: 5381 ident: bib0019 article-title: Digital twin of a combustion furnace operating in flameless conditions: reduced-order model development from CFD simulations publication-title: Proc. Combust. Inst. – volume: 143 year: 2021 ident: bib0004 article-title: Assessment of current capabilities and near-term availability of hydrogen-fired gas turbines considering a low-carbon future publication-title: J. Eng. Gas Turbines Power – volume: 30 start-page: 401 year: 2020 end-page: 429 ident: bib0031 article-title: Data-driven model reduction of multiphase flow in a single-hole automotive injector publication-title: Atomization Sprays – reference: M. Frenklach, C.T. Bowman, G.P. Smith, W.C. Gardiner, World Wide Web location. – volume: 3 start-page: 422 year: 2021 end-page: 440 ident: bib0042 article-title: Physics-informed machine learning publication-title: Nat. Rev. Phys. – volume: 39 start-page: 4541 year: 2023 end-page: 4551 ident: bib0006 article-title: Numerical investigation of boundary layer flashback of CH4/H2/air swirl flames under different thermal boundary conditions in a bluff-body swirl burner publication-title: Proc. Combust. Inst. – volume: 59 start-page: 3291 year: 2021 end-page: 3303 ident: bib0020 article-title: Reduced-order modeling for complex flow emulation by common kernel-smoothed proper orthogonal decomposition publication-title: AIAA J. – volume: 255 year: 2023 ident: bib0005 article-title: Modeling the boundary-layer flashback of premixed hydrogen-enriched swirling flames at high pressures publication-title: Combust. Flame – volume: 57 start-page: 483 year: 2015 end-page: 531 ident: bib0009 article-title: A survey of projection-based model reduction methods for parametric dynamical systems publication-title: SIAM Rev. – volume: 14 start-page: 6480 year: 2023 ident: bib0028 article-title: Grasping extreme aerodynamics on a low-dimensional manifold publication-title: Nat. Commun. – volume: 96 start-page: 23 year: 2018 end-page: 61 ident: bib0043 article-title: A review on design of experiments and surrogate models in aircraft real-time and many-query aerodynamic analyses publication-title: Prog. Aerosp. Sci. – volume: 244 year: 2022 ident: bib0014 article-title: Co-optimized machine-learned manifold models for large eddy simulation of turbulent combustion publication-title: Combust. Flame – volume: 59 start-page: 868 year: 2021 end-page: 879 ident: bib0030 article-title: Machine-learning-based surrogate modeling of aerodynamic flow around distributed structures publication-title: AIAA J. – volume: 341 start-page: 807 year: 2018 end-page: 826 ident: bib0015 article-title: Reduced order modeling for nonlinear structural analysis using Gaussian process regression publication-title: Comput. Methods Appl. Mech. Eng. – volume: 182 start-page: 1 year: 2002 end-page: 26 ident: bib0025 article-title: Neural network modeling for near wall turbulent flow publication-title: J. Comput. Phys. – volume: 259 year: 2020 ident: bib0032 article-title: Combustion stability monitoring through flame imaging and stacked sparse autoencoder based deep neural network publication-title: Appl. Energy – volume: 384 start-page: 289 year: 2019 end-page: 307 ident: bib0011 article-title: Non-intrusive reduced order modeling of unsteady flows using artificial neural networks with application to a combustion problem publication-title: J. Comput. Phys. – volume: 179 start-page: 704 year: 2019 end-page: 717 ident: bib0016 article-title: Projection-based model reduction: formulations for physics-based machine learning publication-title: Comput. Fluids – volume: 393 year: 2022 ident: bib0029 article-title: A comparison of neural network architectures for data-driven reduced-order modeling publication-title: Comput. Methods Appl. Mech. Eng. – volume: 41 start-page: 1 year: 2005 end-page: 28 ident: bib0008 article-title: Surrogate-based analysis and optimization publication-title: Prog. Aerosp. Sci. – volume: 2 start-page: 37 year: 1987 end-page: 52 ident: bib0013 article-title: Principal component analysis publication-title: Chemom. Intell. Lab. Syst. – volume: 249 year: 2023 ident: bib0007 article-title: Investigation on lean blow-off characteristics and stabilization mechanism of premixed hydrogen enhanced ammonia/air swirl flames in a gas turbine combustor publication-title: Combust. Flame – volume: 404 year: 2020 ident: bib0033 article-title: Model reduction of dynamical systems on nonlinear manifolds using deep convolutional autoencoders publication-title: J. Comput. Phys. – volume: 36 year: 2024 ident: bib0021 article-title: Projection-based reduced order modeling of multi-species mixing and combustion publication-title: Phys. Fluids – volume: 25 start-page: 539 year: 1993 end-page: 575 ident: bib0012 article-title: The proper orthogonal decomposition in the analysis of turbulent flows publication-title: Annu. Rev. Fluid Mech. – volume: 113 start-page: 1443 year: 2018 end-page: 1456 ident: bib0039 article-title: An efficient surrogate model for emulation and physics extraction of large eddy simulations publication-title: J. Am. Stat. Assoc. – volume: 32 year: 2020 ident: bib0038 article-title: Convolutional neural network based hierarchical autoencoder for nonlinear mode decomposition of fluid field data publication-title: Phys. Fluids – volume: 96 year: 2023 ident: bib0001 article-title: Biofuels, electrofuels, electric or hydrogen?: A review of current and emerging sustainable aviation systems publication-title: Prog. Energy Combust. Sci. – volume: 38 start-page: 6393 year: 2021 end-page: 6401 ident: bib0018 article-title: Surrogate-based modeling for emulation of supercritical injector flow and combustion publication-title: Proc. Combust. Inst. – volume: 87 start-page: 1 year: 2021 end-page: 36 ident: bib0037 article-title: A comprehensive deep learning-based approach to reduced order modeling of nonlinear time-dependent parametrized PDEs publication-title: J. Sci. Comput. – volume: 168 start-page: 39 year: 2016 end-page: 52 ident: bib0003 article-title: Experimental investigation of upstream flame propagation during boundary layer flashback of swirl flames publication-title: Combust. Flame – volume: 32 year: 2020 ident: bib0023 article-title: Exploration and prediction of fluid dynamical systems using auto-encoder technology publication-title: Phys. Fluids – volume: 52 start-page: 477 year: 2020 end-page: 508 ident: bib0010 article-title: Machine learning for fluid mechanics publication-title: Annu. Rev. Fluid Mech. – volume: 5 year: 1992 ident: bib0024 article-title: Non-linear dimensionality reduction publication-title: Adv. Neural Inf. Process. Syst. – reference: , Version 3.0, 1999. – volume: 372 year: 2020 ident: bib0026 article-title: Multi-level convolutional autoencoder networks for parametric prediction of spatio-temporal dynamics publication-title: Comput. Methods Appl. Mech. Eng. – volume: 9 start-page: 85 year: 2019 end-page: 112 ident: bib0035 article-title: Convolutional neural network: a review of models, methodologies and applications to object detection publication-title: Prog. Artif. Intell. – volume: 37 start-page: 139 year: 2024 end-page: 155 ident: bib0027 article-title: Data-driven surrogate modeling and optimization of supercritical jet into supersonic crossflow publication-title: Chin. J. Aeronaut. – volume: 33 start-page: 6999 year: 2022 end-page: 7019 ident: bib0036 article-title: A survey of convolutional neural networks: analysis, applications, and prospects publication-title: IEEE Trans. Neural Netw. Learn. Syst. – volume: 57 start-page: 5269 year: 2019 end-page: 5280 ident: bib0017 article-title: Kernel-smoothed proper orthogonal decomposition–based emulation for spatiotemporally evolving flow dynamics prediction publication-title: AIAA J. – volume: 3 start-page: 378 year: 2021 end-page: 386 ident: bib0041 article-title: Fitting elephants in modern machine learning by statistically consistent interpolation publication-title: Nat. Mach. Intell. – volume: 313 start-page: 504 year: 2006 end-page: 507 ident: bib0022 article-title: Reducing the dimensionality of data with neural networks publication-title: Science – volume: 262 year: 2024 ident: bib0002 article-title: Hybrid rich- and lean-premixed ammonia-hydrogen combustion for mitigation of NOx emissions and thermoacoustic instabilities publication-title: Combust. Flame – volume: 252 year: 2023 ident: bib0034 article-title: 3-D soot temperature and volume fraction reconstruction of afterburner flame via deep learning algorithms publication-title: Combust. Flame – volume: 143 year: 2021 ident: 10.1016/j.combustflame.2025.113981_bib0004 article-title: Assessment of current capabilities and near-term availability of hydrogen-fired gas turbines considering a low-carbon future publication-title: J. Eng. Gas Turbines Power doi: 10.1115/1.4049346 – volume: 244 year: 2022 ident: 10.1016/j.combustflame.2025.113981_bib0014 article-title: Co-optimized machine-learned manifold models for large eddy simulation of turbulent combustion publication-title: Combust. Flame doi: 10.1016/j.combustflame.2022.112286 – volume: 182 start-page: 1 year: 2002 ident: 10.1016/j.combustflame.2025.113981_bib0025 article-title: Neural network modeling for near wall turbulent flow publication-title: J. Comput. Phys. doi: 10.1006/jcph.2002.7146 – volume: 38 start-page: 6393 year: 2021 ident: 10.1016/j.combustflame.2025.113981_bib0018 article-title: Surrogate-based modeling for emulation of supercritical injector flow and combustion publication-title: Proc. Combust. Inst. doi: 10.1016/j.proci.2020.06.303 – volume: 59 start-page: 868 year: 2021 ident: 10.1016/j.combustflame.2025.113981_bib0030 article-title: Machine-learning-based surrogate modeling of aerodynamic flow around distributed structures publication-title: AIAA J. doi: 10.2514/1.J059877 – volume: 252 year: 2023 ident: 10.1016/j.combustflame.2025.113981_bib0034 article-title: 3-D soot temperature and volume fraction reconstruction of afterburner flame via deep learning algorithms publication-title: Combust. Flame doi: 10.1016/j.combustflame.2023.112743 – volume: 30 start-page: 401 year: 2020 ident: 10.1016/j.combustflame.2025.113981_bib0031 article-title: Data-driven model reduction of multiphase flow in a single-hole automotive injector publication-title: Atomization Sprays doi: 10.1615/AtomizSpr.2020034830 – volume: 393 year: 2022 ident: 10.1016/j.combustflame.2025.113981_bib0029 article-title: A comparison of neural network architectures for data-driven reduced-order modeling publication-title: Comput. Methods Appl. Mech. Eng. doi: 10.1016/j.cma.2022.114764 – volume: 96 start-page: 23 year: 2018 ident: 10.1016/j.combustflame.2025.113981_bib0043 article-title: A review on design of experiments and surrogate models in aircraft real-time and many-query aerodynamic analyses publication-title: Prog. Aerosp. Sci. doi: 10.1016/j.paerosci.2017.11.003 – volume: 262 year: 2024 ident: 10.1016/j.combustflame.2025.113981_bib0002 article-title: Hybrid rich- and lean-premixed ammonia-hydrogen combustion for mitigation of NOx emissions and thermoacoustic instabilities publication-title: Combust. Flame doi: 10.1016/j.combustflame.2024.113366 – volume: 38 start-page: 5373 year: 2021 ident: 10.1016/j.combustflame.2025.113981_bib0019 article-title: Digital twin of a combustion furnace operating in flameless conditions: reduced-order model development from CFD simulations publication-title: Proc. Combust. Inst. doi: 10.1016/j.proci.2020.06.045 – volume: 59 start-page: 3291 year: 2021 ident: 10.1016/j.combustflame.2025.113981_bib0020 article-title: Reduced-order modeling for complex flow emulation by common kernel-smoothed proper orthogonal decomposition publication-title: AIAA J. doi: 10.2514/1.J060574 – volume: 14 start-page: 6480 year: 2023 ident: 10.1016/j.combustflame.2025.113981_bib0028 article-title: Grasping extreme aerodynamics on a low-dimensional manifold publication-title: Nat. Commun. doi: 10.1038/s41467-023-42213-6 – volume: 168 start-page: 39 year: 2016 ident: 10.1016/j.combustflame.2025.113981_bib0003 article-title: Experimental investigation of upstream flame propagation during boundary layer flashback of swirl flames publication-title: Combust. Flame doi: 10.1016/j.combustflame.2016.03.027 – volume: 372 year: 2020 ident: 10.1016/j.combustflame.2025.113981_bib0026 article-title: Multi-level convolutional autoencoder networks for parametric prediction of spatio-temporal dynamics publication-title: Comput. Methods Appl. Mech. Eng. doi: 10.1016/j.cma.2020.113379 – ident: 10.1016/j.combustflame.2025.113981_bib0040 – volume: 57 start-page: 5269 year: 2019 ident: 10.1016/j.combustflame.2025.113981_bib0017 article-title: Kernel-smoothed proper orthogonal decomposition–based emulation for spatiotemporally evolving flow dynamics prediction publication-title: AIAA J. doi: 10.2514/1.J057803 – volume: 37 start-page: 139 year: 2024 ident: 10.1016/j.combustflame.2025.113981_bib0027 article-title: Data-driven surrogate modeling and optimization of supercritical jet into supersonic crossflow publication-title: Chin. J. Aeronaut. doi: 10.1016/j.cja.2024.08.012 – volume: 179 start-page: 704 year: 2019 ident: 10.1016/j.combustflame.2025.113981_bib0016 article-title: Projection-based model reduction: formulations for physics-based machine learning publication-title: Comput. Fluids doi: 10.1016/j.compfluid.2018.07.021 – volume: 96 year: 2023 ident: 10.1016/j.combustflame.2025.113981_bib0001 article-title: Biofuels, electrofuels, electric or hydrogen?: A review of current and emerging sustainable aviation systems publication-title: Prog. Energy Combust. Sci. doi: 10.1016/j.pecs.2023.101073 – volume: 3 start-page: 422 year: 2021 ident: 10.1016/j.combustflame.2025.113981_bib0042 article-title: Physics-informed machine learning publication-title: Nat. Rev. Phys. doi: 10.1038/s42254-021-00314-5 – volume: 25 start-page: 539 year: 1993 ident: 10.1016/j.combustflame.2025.113981_bib0012 article-title: The proper orthogonal decomposition in the analysis of turbulent flows publication-title: Annu. Rev. Fluid Mech. doi: 10.1146/annurev.fl.25.010193.002543 – volume: 404 year: 2020 ident: 10.1016/j.combustflame.2025.113981_bib0033 article-title: Model reduction of dynamical systems on nonlinear manifolds using deep convolutional autoencoders publication-title: J. Comput. Phys. doi: 10.1016/j.jcp.2019.108973 – volume: 5 year: 1992 ident: 10.1016/j.combustflame.2025.113981_bib0024 article-title: Non-linear dimensionality reduction publication-title: Adv. Neural Inf. Process. Syst. – volume: 341 start-page: 807 year: 2018 ident: 10.1016/j.combustflame.2025.113981_bib0015 article-title: Reduced order modeling for nonlinear structural analysis using Gaussian process regression publication-title: Comput. Methods Appl. Mech. Eng. doi: 10.1016/j.cma.2018.07.017 – volume: 33 start-page: 6999 year: 2022 ident: 10.1016/j.combustflame.2025.113981_bib0036 article-title: A survey of convolutional neural networks: analysis, applications, and prospects publication-title: IEEE Trans. Neural Netw. Learn. Syst. doi: 10.1109/TNNLS.2021.3084827 – volume: 87 start-page: 1 year: 2021 ident: 10.1016/j.combustflame.2025.113981_bib0037 article-title: A comprehensive deep learning-based approach to reduced order modeling of nonlinear time-dependent parametrized PDEs publication-title: J. Sci. Comput. doi: 10.1007/s10915-021-01462-7 – volume: 2 start-page: 37 year: 1987 ident: 10.1016/j.combustflame.2025.113981_bib0013 article-title: Principal component analysis publication-title: Chemom. Intell. Lab. Syst. doi: 10.1016/0169-7439(87)80084-9 – volume: 255 year: 2023 ident: 10.1016/j.combustflame.2025.113981_bib0005 article-title: Modeling the boundary-layer flashback of premixed hydrogen-enriched swirling flames at high pressures publication-title: Combust. Flame doi: 10.1016/j.combustflame.2023.112900 – volume: 32 year: 2020 ident: 10.1016/j.combustflame.2025.113981_bib0038 article-title: Convolutional neural network based hierarchical autoencoder for nonlinear mode decomposition of fluid field data publication-title: Phys. Fluids doi: 10.1063/5.0020721 – volume: 41 start-page: 1 year: 2005 ident: 10.1016/j.combustflame.2025.113981_bib0008 article-title: Surrogate-based analysis and optimization publication-title: Prog. Aerosp. Sci. doi: 10.1016/j.paerosci.2005.02.001 – volume: 113 start-page: 1443 year: 2018 ident: 10.1016/j.combustflame.2025.113981_bib0039 article-title: An efficient surrogate model for emulation and physics extraction of large eddy simulations publication-title: J. Am. Stat. Assoc. doi: 10.1080/01621459.2017.1409123 – volume: 313 start-page: 504 year: 2006 ident: 10.1016/j.combustflame.2025.113981_bib0022 article-title: Reducing the dimensionality of data with neural networks publication-title: Science doi: 10.1126/science.1127647 – volume: 32 year: 2020 ident: 10.1016/j.combustflame.2025.113981_bib0023 article-title: Exploration and prediction of fluid dynamical systems using auto-encoder technology publication-title: Phys. Fluids doi: 10.1063/5.0012906 – volume: 249 year: 2023 ident: 10.1016/j.combustflame.2025.113981_bib0007 article-title: Investigation on lean blow-off characteristics and stabilization mechanism of premixed hydrogen enhanced ammonia/air swirl flames in a gas turbine combustor publication-title: Combust. Flame doi: 10.1016/j.combustflame.2022.112600 – volume: 39 start-page: 4541 year: 2023 ident: 10.1016/j.combustflame.2025.113981_bib0006 article-title: Numerical investigation of boundary layer flashback of CH4/H2/air swirl flames under different thermal boundary conditions in a bluff-body swirl burner publication-title: Proc. Combust. Inst. doi: 10.1016/j.proci.2022.07.040 – volume: 9 start-page: 85 year: 2019 ident: 10.1016/j.combustflame.2025.113981_bib0035 article-title: Convolutional neural network: a review of models, methodologies and applications to object detection publication-title: Prog. Artif. Intell. doi: 10.1007/s13748-019-00203-0 – volume: 3 start-page: 378 year: 2021 ident: 10.1016/j.combustflame.2025.113981_bib0041 article-title: Fitting elephants in modern machine learning by statistically consistent interpolation publication-title: Nat. Mach. Intell. doi: 10.1038/s42256-021-00345-8 – volume: 57 start-page: 483 year: 2015 ident: 10.1016/j.combustflame.2025.113981_bib0009 article-title: A survey of projection-based model reduction methods for parametric dynamical systems publication-title: SIAM Rev. doi: 10.1137/130932715 – volume: 36 year: 2024 ident: 10.1016/j.combustflame.2025.113981_bib0021 article-title: Projection-based reduced order modeling of multi-species mixing and combustion publication-title: Phys. Fluids doi: 10.1063/5.0217845 – volume: 52 start-page: 477 year: 2020 ident: 10.1016/j.combustflame.2025.113981_bib0010 article-title: Machine learning for fluid mechanics publication-title: Annu. Rev. Fluid Mech. doi: 10.1146/annurev-fluid-010719-060214 – volume: 384 start-page: 289 year: 2019 ident: 10.1016/j.combustflame.2025.113981_bib0011 article-title: Non-intrusive reduced order modeling of unsteady flows using artificial neural networks with application to a combustion problem publication-title: J. Comput. Phys. doi: 10.1016/j.jcp.2019.01.031 – volume: 259 year: 2020 ident: 10.1016/j.combustflame.2025.113981_bib0032 article-title: Combustion stability monitoring through flame imaging and stacked sparse autoencoder based deep neural network publication-title: Appl. Energy doi: 10.1016/j.apenergy.2019.114159 |
| SSID | ssj0007433 |
| Score | 2.498305 |
| Snippet | Numerical simulations are essential in comprehending the processes of flow and combustion in aerospace and power-generation systems. Yet, substantial... |
| SourceID | crossref elsevier |
| SourceType | Enrichment Source Index Database Publisher |
| StartPage | 113981 |
| SubjectTerms | Convolutional autoencoder Kriging Nonlinear feature Proper orthogonal decomposition Reduced-order model (ROM) |
| Title | Reduced-order modeling via convolutional autoencoder for emulating combustion of hydrogen/methane fuel blends |
| URI | https://dx.doi.org/10.1016/j.combustflame.2025.113981 |
| Volume | 274 |
| WOSCitedRecordID | wos001416272900001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| journalDatabaseRights | – providerCode: PRVESC databaseName: ScienceDirect issn: 0010-2180 databaseCode: AIEXJ dateStart: 19950101 customDbUrl: isFulltext: true dateEnd: 99991231 titleUrlDefault: https://www.sciencedirect.com omitProxy: false ssIdentifier: ssj0007433 providerName: Elsevier |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV3fb9MwELbKhgQ8IBggBgP5gTeUkTZ1Ez_wMHWbBpomYAP1LbIde-vUJVNpqo6_njv_SNMBUhHiJWpPvbr1fXXPzt33EfJGpbKrTCaiWCgW9SVDDsg0iSRjWqNWErMd3t-O05OTbDTinzodE3ph5pO0LLPFgl__11CDDYKNrbN_Ee7mTcEAjyHocIWww3WtwH9BMlZdRJZU0ynd4HHAfCxsibkfGykC6lmFNJb4Mkv8fWWlvFwTrkSVL59L3hTT6hyXp0PUmxaQlppaT97KifZtwg3VwdIPz-MNoK3Bzb4XTzkd39TLmyHujr8uF41teFGjcdQYju3zz-D-46Kq26cUPdYqbrFHZ6F9ZqW6EzekEeQYcXs57jnVnl-WdnfKcLnrp8B-hV0cCmVpuNN9uUWdfWoZ85DjjtlcNr1DNnsp47CAb-59OBh9bP6zIY9ytQj-AwV6WlsJ-KcRf5_KtNKTs0fkod9X0D2Hh8eko8stcm8Y5Py2yIMW8-QTcrWCEhpQQgEldAUltIUSCiihDUroEiW0MjSg5J3HCEWMUIeRp-Tr4cHZ8CjyyhuRggR7FgkkaoyTRMSZgQ2pUEYXA81FPBC4hUiRQQjmqkCqKMYyOeDS6KSfMRVL0YOM-hnZKKtSPycUHGWSMGESrvqMSQEbcmO6kneLfiaF3iY8TGKuPC09qqNM8lB_eJm3A5BjAHIXgG2SNL7XjpxlLa_3IVa5TzNd-pgD1Nbwf_GP_i_J_eUvZIdszKa1fkXuqvls_H362iPzJ1S4rtk |
| linkProvider | Elsevier |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Reduced-order+modeling+via+convolutional+autoencoder+for+emulating+combustion+of+hydrogen%2Fmethane+fuel+blends&rft.jtitle=Combustion+and+flame&rft.au=Ding%2C+Siyu&rft.au=Ni%2C+Chenxu&rft.au=Chu%2C+Xu&rft.au=Lu%2C+Qingzhou&rft.date=2025-04-01&rft.pub=Elsevier+Inc&rft.issn=0010-2180&rft.volume=274&rft_id=info:doi/10.1016%2Fj.combustflame.2025.113981&rft.externalDocID=S0010218025000197 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0010-2180&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0010-2180&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0010-2180&client=summon |