Article(id=1315732079397163745, tenantId=1146029695717560320, journalId=1315122079294652549, issueId=1315732078327616225, articleNumber=null, orderNo=null, doi=10.11883/bzycj-2025-0218, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1752422400000, receivedDateStr=2025-07-14, revisedDate=1774627200000, revisedDateStr=2026-03-28, acceptedDate=null, acceptedDateStr=null, onlineDate=1791624572847, onlineDateStr=2026-10-10, pubDate=1788537600000, pubDateStr=2026-09-05, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1791624572847, onlineIssueDateStr=2026-10-10, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1791624572847, creator=13701087609, updateTime=1791624572847, updator=13701087609, issue=Issue{id=1315732078327616225, tenantId=1146029695717560320, journalId=1315122079294652549, year='2026', volume='46', issue='9', pageStart='091001-1', pageEnd='095402-10', issueExtLink='null', onlineDate='null', pubDate='1788537600000', pubDateStr='2026-09-05', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=1, specialIssue=null, createTime=1791624572593, creator='13701087609', updateTime=1791624783658, updator='13701087609', preIssue=null, nextIssue=null, articleTotal=null, ext={EN=IssueExt(id=1315732963703247483, tenantId=1146029695717560320, journalId=1315122079294652549, issueId=1315732078327616225, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1315732963703247484, tenantId=1146029695717560320, journalId=1315122079294652549, issueId=1315732078327616225, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null, downloadFileDto=null}, startPage=091502-1, endPage=091502-12, ext={EN=ArticleExt(id=1315732079665599203, articleId=1315732079397163745, tenantId=1146029695717560320, journalId=1315122079294652549, language=EN, title=Prediction of critical quenching diameter for deflagration and detonation based on deep learning algorithms, columnId=1315732079569130210, journalTitle=Explosion and Shock Waves, columnName=Interdisciplinary Frontiers, runingTitle=null, highlight=null, articleAbstract=

The research addresses the safety imperative of preventing flammable gas explosions in enclosed pipelines by establishing a predictive model for the critical quenching diameter within porous media flame arresters. A novel predictive framework was developed based on a comprehensive nine-dimensional feature space incorporating gas composition parameters (e.g., hydrogen equivalence ratio), pipeline geometry dimensions (length-to-diameter ratio), initial thermodynamic conditions (pressure), and porous medium structural characteristics (thickness, material thermal conductivity). A systematic investigation was conducted to identify the optimal hyperparameter configurations for both convolutional neural network (CNN) and Transformer architectures. Rigorous validation demonstrated that the Transformer model exhibits statistically significant superiority over the CNN model across all key performance metrics. Specifically, the model achieved a δMAE (mean absolute error, MAE) of 0.068, a δMSE (mean squared error, MSE) of 0.008, and an coefficient of determination R2 of 0.928. The performance notably surpassed the CNN results (δMAE=0.079, δMSE=0.012, R2=0.906). Beyond evaluation indicators, detailed error distribution analysis confirmed the Transformer's enhanced predictive accuracy and reduced susceptibility to outliers. The superior performance is attributed to the Transformer’s intrinsic self-attention mechanism, which excels at dynamically identifying and weighting critical interdependencies among the diverse input features governing the complex quenching process. The capability enables more precise capture of the nonlinear phenomena defining the quenching limit. Furthermore, robustness testing involving diverse data normalization schemes revealed that the Transformer model exhibits greater stability. This resilience stems from its inherent layer normalization mechanism, which effectively decouples feature dependencies and mitigates sensitivity to input scaling variations. Consequently, the Transformer architecture is established as the definitive optimal model for this critical safety prediction task. Its significant advantage lies in the minimal data preprocessing required before deployment, a feature that greatly enhances its practical utility. The model provides robust quantitative decision-making support essential for formulating effective gas explosion mitigation strategies and optimizing the safety design parameters of pipeline flame arresters. By accurately predicting the critical quenching diameter across various scenarios, this work delivers a valuable tool for enhancing inherent safety in industries handling combustible gases within confined pipeline systems.

, authors=Zhongheng NIE1, 2, Li WANG2, Wei GAO1, Haipeng JIANG1, authorsList=Zhongheng NIE, Li WANG, Wei GAO, Haipeng JIANG, authorCompany=null, correspAuthors=Haipeng JIANG, authorNote=null, correspAuthorsNote=null, copyrightStatement=Explosion and Shock Waves, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=null, magXml=null, pdfUrl=null, pdf=null, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=null, mapNumber=null, fund=null), CN=ArticleExt(id=1315732083855708922, articleId=1315732079397163745, tenantId=1146029695717560320, journalId=1315122079294652549, language=CN, title=基于深度学习算法预测燃爆的临界淬熄直径, columnId=1315732079833371364, journalTitle=爆炸与冲击, columnName=学科交叉前沿, runingTitle=null, highlight=null, articleAbstract=

聚焦密闭管道可燃气体燃爆火焰淬熄的安全防护需求,基于燃气组分、管道几何、初始条件及多孔介质结构参数构建了九维特征空间,建立了多孔介质临界淬熄直径预测模型。为突破传统经验公式的精度不足问题与应用局限性,通过系统性的超参数优化与模型验证,深入对比并证实了Transformer架构在临界淬熄直径预测问题上的显著优越性:其预测性能(平均绝对误差δMAE=0.068,均方误差δMSE=0.008,相关系数R2=0.928)显著超越了广泛应用的卷积神经网络(convolutional neural network,CNN)模型(δMAE=0.079,δMSE=0.012,R2=0.906),不仅整体误差更低,且对离群点的鲁棒性更强。深入分析发现,Transformer模型的核心优势源于其自注意力机制对淬熄过程中关键临界特征的精准捕获与高效建模能力。在数据归一化敏感性验证中,Transformer模型展现出优异的鲁棒性,这归功于其层归一化机制所赋予的特征解耦与稳定表示能力。基于上述系统性评估,最终确立Transformer模型为预测多孔介质临界淬熄直径的最优模型,可为燃爆安全防控策略的量化制定及管道阻火器安全性能的精细化设计提供强大的、可操作的决策支持工具,具有重要的理论指导意义。

, authors=聂仲恒1, 2, 汪丽2, 高伟1, 姜海鹏1, authorsList=聂仲恒, 汪丽, 高伟, 姜海鹏, authorCompany=null, correspAuthors=姜海鹏, authorNote=

聂仲恒(2000- ),男,硕士,

, correspAuthorsNote=
姜海鹏(1990- ),男,博士,副教授,
, copyrightStatement=版权所有©《爆炸与冲击》编辑部 2026, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=kXwVUsZ2bHKWVjnMXfxb4A==, magXml=13ahsjKm6HH1zYY4WRr24g==, pdfUrl=null, pdf=HUCnezQ3f1a57q+/x3VwKQ==, pdfFileSize=1810220, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=DuSkmUyql8UVf7MyFkKzog==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=iV+Tl6g+tGkiHahvtOTE4w==, mapNumber=null, fund=null)}, authors=[Author(id=1315732084233196290, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=nzh2490823579@163.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1315732084300305157, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, authorId=1315732084233196290, language=EN, stringName=Zhongheng NIE, firstName=Zhongheng, middleName=null, lastName=NIE, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1Dalian University of Technology, Dalian 116024, Liaoning, China
2Beijing Institute of Aerospace Testing Technology, Beijing 100074, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1315732084371608326, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, authorId=1315732084233196290, language=CN, stringName=聂仲恒, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1大连理工大学化工学院,辽宁省 大连市 116024
2北京航天试验技术研究所,北京 100074, bio={"content":"

聂仲恒(2000- ),男,硕士,

"}, bioImg=null, bioContent=

聂仲恒(2000- ),男,硕士,

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1315732084048646907, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, xref=1, ext=[AuthorCompanyExt(id=1315732084057035516, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, companyId=1315732084048646907, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1Dalian University of Technology, Dalian 116024, Liaoning, China), AuthorCompanyExt(id=1315732084065424125, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, companyId=1315732084048646907, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1大连理工大学化工学院,辽宁省 大连市 116024)]), AuthorCompany(id=1315732084157698814, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, xref=2, ext=[AuthorCompanyExt(id=1315732084161893119, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, companyId=1315732084157698814, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Beijing Institute of Aerospace Testing Technology, Beijing 100074, China), AuthorCompanyExt(id=1315732084170281728, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, companyId=1315732084157698814, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2北京航天试验技术研究所,北京 100074)])]), Author(id=1315732084438717192, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1315732084510020362, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, authorId=1315732084438717192, language=EN, stringName=Li WANG, firstName=Li, middleName=null, lastName=WANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2Beijing Institute of Aerospace Testing Technology, Beijing 100074, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1315732084585517835, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, authorId=1315732084438717192, language=CN, stringName=汪丽, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2北京航天试验技术研究所,北京 100074, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1315732084157698814, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, xref=2, ext=[AuthorCompanyExt(id=1315732084161893119, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, companyId=1315732084157698814, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Beijing Institute of Aerospace Testing Technology, Beijing 100074, China), AuthorCompanyExt(id=1315732084170281728, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, companyId=1315732084157698814, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2北京航天试验技术研究所,北京 100074)])]), Author(id=1315732084652626701, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1315732084728124175, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, authorId=1315732084652626701, language=EN, stringName=Wei GAO, firstName=Wei, middleName=null, lastName=GAO, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1Dalian University of Technology, Dalian 116024, Liaoning, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1315732084803621648, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, authorId=1315732084652626701, language=CN, stringName=高伟, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1大连理工大学化工学院,辽宁省 大连市 116024, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1315732084048646907, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, xref=1, ext=[AuthorCompanyExt(id=1315732084057035516, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, companyId=1315732084048646907, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1Dalian University of Technology, Dalian 116024, Liaoning, China), AuthorCompanyExt(id=1315732084065424125, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, companyId=1315732084048646907, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1大连理工大学化工学院,辽宁省 大连市 116024)])]), Author(id=1315732084883313426, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=jhp@dlut.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1315732084963005204, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, authorId=1315732084883313426, language=EN, stringName=Haipeng JIANG, firstName=Haipeng, middleName=null, lastName=JIANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1Dalian University of Technology, Dalian 116024, Liaoning, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1315732085030114069, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, authorId=1315732084883313426, language=CN, stringName=姜海鹏, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1大连理工大学化工学院,辽宁省 大连市 116024, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1315732084048646907, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, xref=1, ext=[AuthorCompanyExt(id=1315732084057035516, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, companyId=1315732084048646907, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1Dalian University of Technology, Dalian 116024, Liaoning, China), AuthorCompanyExt(id=1315732084065424125, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, companyId=1315732084048646907, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1大连理工大学化工学院,辽宁省 大连市 116024)])])], keywords=[Keyword(id=1315732085172720406, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=EN, orderNo=1, keyword=pipeline explosion), Keyword(id=1315732085235634967, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=EN, orderNo=2, keyword=deep learning), Keyword(id=1315732085298549528, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=EN, orderNo=3, keyword=critical quenching diameter prediction), Keyword(id=1315732085365658393, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=EN, orderNo=4, keyword=flame arrester), Keyword(id=1315732085436961562, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=CN, orderNo=1, keyword=管道爆炸), Keyword(id=1315732085504070427, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=CN, orderNo=2, keyword=深度学习), Keyword(id=1315732085571179292, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=CN, orderNo=3, keyword=临界淬熄直径预测), Keyword(id=1315732087206957853, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=CN, orderNo=4, keyword=阻火器)], refs=[Reference(id=1315732092630192955, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=null, journalName=null, refType=null, unstructuredReference=GROSSEL S S. Deflagration and detonation flame arresters [M]. America: Center for Chemical Process Safety of the American Institute of Chemical Engineers, 2002., articleTitle=null, refAbstract=null), Reference(id=1315732092823130940, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[1], rfOrder=1, authorNames=null, journalName=null, refType=null, unstructuredReference=GROSSEL S S. Deflagration and detonation flame arresters [M]. America: Center for Chemical Process Safety of the American Institute of Chemical Engineers, 2002. DOI: 10.1002/9780470935651., articleTitle=null, refAbstract=null), Reference(id=1315732092894434109, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[2], rfOrder=2, authorNames=null, journalName=null, refType=null, unstructuredReference=NIE Z H, GAO W, JIANG H P, et al. Predicting critical flame quenching thickness using machine learning approach with ResNet and ANN [J]. Journal of Loss Prevention in the Process Industries, 2024, 92: 105448. DOI: 10.1016/j.jlp.2024.105448., articleTitle=null, refAbstract=null), Reference(id=1315732092995097406, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[3], rfOrder=3, authorNames=null, journalName=null, refType=null, unstructuredReference=LANGFORD B, PALMER K N, TONKIN P S. The performance of flame arresters against flames propagating in various fuel/air mixtures: Fire Research Note 486 [R]. Herts: Fire Research Station, 1961., articleTitle=null, refAbstract=null), Reference(id=1315732093066400575, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[4], rfOrder=4, authorNames=null, journalName=null, refType=null, unstructuredReference=SONG X Z, ZUO X C, YANG Z K, et al. The explosion-suppression performance of mesh aluminum alloys and spherical nonmetallic materials on hydrogen-air mixtures [J]. International Journal of Hydrogen Energy, 2020, 45(56): 32686–32701. DOI: 10.1016/j.ijhydene.2020.08.197., articleTitle=null, refAbstract=null), Reference(id=1315732093120926528, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[5], rfOrder=5, authorNames=null, journalName=null, refType=null, unstructuredReference=NIE B S, HE X Q, ZHANG R M, et al. The roles of foam ceramics in suppression of gas explosion overpressure and quenching of flame propagation [J]. Journal of Hazardous Materials, 2011, 192(2): 741–747. DOI: 10.1016/j.jhazmat.2011.05.083., articleTitle=null, refAbstract=null), Reference(id=1315732093188035393, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[6], rfOrder=6, authorNames=null, journalName=null, refType=null, unstructuredReference=SUN S C, LIU G, LIU J X, et al. Effect of porosity and element thickness on flame quenching for in-line crimped-ribbon flame arresters [J]. Journal of Loss Prevention in the Process Industries, 2017, 50: 221–228. DOI: 10.1016/j.jlp.2017.09.017., articleTitle=null, refAbstract=null), Reference(id=1315732093267727170, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[7], rfOrder=7, authorNames=null, journalName=null, refType=null, unstructuredReference=CAO X Y, WANG Z, WANG Y, et al. Effect of wire mesh structure parameter on the flame propagation characteristics of syngas explosion [J]. Fuel, 2023, 334: 126658. DOI: 10.1016/j.fuel.2022.126658., articleTitle=null, refAbstract=null), Reference(id=1315732093339030339, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[8], rfOrder=8, authorNames=null, journalName=null, refType=null, unstructuredReference=LIU Z Y, GU Z Y, LIU P. Deep learning [M]//LIU Z Y, GU Z Y, LIU P. Transportation Big Data: Theory and Methods. Amsterdam: Elsevier, 2025: 383–430., articleTitle=null, refAbstract=null), Reference(id=1315732093401944900, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[9], rfOrder=9, authorNames=null, journalName=null, refType=null, unstructuredReference=ZHAO F Y, GAO W, LU J X, et al. Real-time concentration detection of Al dust using GRU-based Kalman filtering approach [J]. Process Safety and Environmental Protection, 2024, 189: 154–163. DOI: 10.1016/j.psep.2024.06.052., articleTitle=null, refAbstract=null), Reference(id=1315732093473248069, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[10], rfOrder=10, authorNames=null, journalName=null, refType=null, unstructuredReference=ZHANG K, ZHANG K, BAO R. Prediction of gas explosion pressures: a machine learning algorithm based on KPCA and an optimized LSSVM [J]. Journal of Loss Prevention in the Process Industries, 2023, 83: 105082. DOI: 10.1016/j.jlp.2023.105082., articleTitle=null, refAbstract=null), Reference(id=1315732093544551238, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[11], rfOrder=11, authorNames=null, journalName=null, refType=null, unstructuredReference=CHEN J G, NIE Z H, ZHAO F Y, et al. Improving the stability of electrostatic induction dust concentration detection using kalman filtering algorithm aided by machine learning [J]. Process Safety and Environmental Protection, 2023, 174: 882–890. DOI: 10.1016/j.psep.2023.05.013., articleTitle=null, refAbstract=null), Reference(id=1315732093611660103, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[12], rfOrder=12, authorNames=null, journalName=null, refType=null, unstructuredReference=GUIBERTI T F, BELHI M, DAMAZO J S, et al. Quenching distance of laminar methane-air flames at cryogenic temperatures and implications for flame arrester design [J]. Applications in Energy and Combustion Science, 2020, 1: 100001. DOI: 10.1016/j.jaecs.2020.100001., articleTitle=null, refAbstract=null), Reference(id=1315732093703934792, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[13], rfOrder=13, authorNames=null, journalName=null, refType=null, unstructuredReference=FRIEDMAN R. The quenching of laminar oxyhydrogen flames by solid surfaces [J]. Symposium on Combustion and Flame, and Explosion Phenomena, 1948, 3(1): 110–120. DOI: 10.1016/S1062-2896(49)80014-6., articleTitle=null, refAbstract=null), Reference(id=1315732093775237961, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[14], rfOrder=14, authorNames=null, journalName=null, refType=null, unstructuredReference=WEN X P, MA M, SUN W C, et al. The quenching characteristics of gas deflagration flame in narrow channel [J]. Advanced Materials Research, 2012, 455/456: 289–295. DOI: 10.4028/www.scientific.net/AMR.455-456.289., articleTitle=null, refAbstract=null), Reference(id=1315732093854929738, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[15], rfOrder=15, authorNames=null, journalName=null, refType=null, unstructuredReference=WU Y, WEN X P, GUO Z D, et al. Experimental study on the propagation characteristics of hydrogen/methane/air premixed flames in a narrow channel [J]. International Journal of Hydrogen Energy, 2022, 47(9): 6377–6387. DOI: 10.1016/j.ijhydene.2021.12.019., articleTitle=null, refAbstract=null), Reference(id=1315732093905261387, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[16], rfOrder=16, authorNames=null, journalName=null, refType=null, unstructuredReference=LONG F Y, DUAN Y L, YU S W, et al. Effect of porous materials on explosion characteristics of low ratio hydrogen/methane mixture in barrier tube [J]. Journal of Loss Prevention in the Process Industries, 2022, 80: 104875. DOI: 10.1016/j.jlp.2022.104875., articleTitle=null, refAbstract=null), Reference(id=1315732093972370252, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[17], rfOrder=17, authorNames=null, journalName=null, refType=null, unstructuredReference=周凯元 李宗芬, 周自金. 波纹板阻火器对爆燃火焰淬熄作用的实验研究 [J]. 中国科学技术大学学报, 1997, 27(4): 77–82., articleTitle=null, refAbstract=null), Reference(id=1315732094035284813, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[17], rfOrder=18, authorNames=null, journalName=null, refType=null, unstructuredReference=ZHOU K Y, LI Z F, ZHOU Z J. The quenching of deflagration by crimped-ribbon flame arresters [J]. Journal of China University of Science and Technology, 1997, 27(4): 77–82., articleTitle=null, refAbstract=null), Reference(id=1315732094094005070, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[18], rfOrder=19, authorNames=null, journalName=null, refType=null, unstructuredReference=WANG L Q, MA H H, SHEN Z W. The quenching of propane deflagrations by crimped ribbon flame arrestors [J]. Journal of Loss Prevention in the Process Industries, 2016, 43: 567–574. DOI: 10.1016/j.jlp.2016.07.025., articleTitle=null, refAbstract=null), Reference(id=1315732094161113935, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[19], rfOrder=20, authorNames=null, journalName=null, refType=null, unstructuredReference=HOLM J M. II. On the initiation of gaseous explosions by small flames [J]. The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science, 1932, 14(89): 18–56. DOI: 10.1080/14786443209462032., articleTitle=null, refAbstract=null), Reference(id=1315732094232417104, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[20], rfOrder=21, authorNames=null, journalName=null, refType=null, unstructuredReference=PIOTROWSKI T C. Specification of flame arresting devices for manifolded low pressure storage tanks [J]. Plant/operations Progress, 1991, 10(2): 102–106. DOI: 10.1002/prsb.720100210., articleTitle=null, refAbstract=null), Reference(id=1315732094295331665, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[21], rfOrder=22, authorNames=null, journalName=null, refType=null, unstructuredReference=CHEN D G, MA H H, SHEN Z W, et al. Investigation of the quenching of methane-air deflagration in narrow parallel channels [J]. Process Safety Progress, 2020, 39(3): e12126. DOI: 10.1002/prs.12126., articleTitle=null, refAbstract=null), Reference(id=1315732094362440530, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[22], rfOrder=23, authorNames=null, journalName=null, refType=null, unstructuredReference=LV X S, YU J L, HOU Y J, et al. The quenching and attenuation of hydrogen-air detonation after passing across capillaries [J]. Fuel, 2022, 324: 124535. DOI: 10.1016/j.fuel.2022.124535., articleTitle=null, refAbstract=null), Reference(id=1315732094429549395, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[23], rfOrder=24, authorNames=null, journalName=null, refType=null, unstructuredReference=SHI J H, ZHANG H, LI J J, et al. Real-time natural gas explosion modeling of offshore platforms by using deep learning probability approach [J]. Ocean Engineering, 2023, 276: 114244. DOI: 10.1016/j.oceaneng.2023.114244., articleTitle=null, refAbstract=null), Reference(id=1315732094505046868, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[24], rfOrder=25, authorNames=null, journalName=null, refType=null, unstructuredReference=SUZUKI Y, NAKAYAMA J, SUZUKI T, et al. Consequence analysis of a small-scale hydrogen leakage from the overhead hydrogen piping based on machine learning and physical modeling [J]. Journal of Loss Prevention in the Process Industries, 2024, 90: 105328. DOI: 10.1016/j.jlp.2024.105328., articleTitle=null, refAbstract=null), Reference(id=1315732096140825429, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[25], rfOrder=26, authorNames=null, journalName=null, refType=null, unstructuredReference=QIN G J, XIA A L, LU H F, et al. A hybrid machine learning model for predicting crater width formed by explosions of natural gas pipelines [J]. Journal of Loss Prevention in the Process Industries, 2023, 82: 104994. DOI: 10.1016/j.jlp.2023.104994., articleTitle=null, refAbstract=null), Reference(id=1315732096245683030, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[26], rfOrder=27, authorNames=null, journalName=null, refType=null, unstructuredReference=XU Q M, CHEN G H, SU S, et al. Prediction of venting gas explosion overpressure based on a combination of explosive theory and machine learning [J]. Expert Systems with Applications, 2023, 234: 121044. DOI: 10.1016/j.eswa.2023.121044., articleTitle=null, refAbstract=null), Reference(id=1315732096363123543, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[27], rfOrder=28, authorNames=null, journalName=null, refType=null, unstructuredReference=NIE Z H, GAO W, JIANG H P, et al. Quenching characteristics and mechanism of hydrogen-air mixtures by corrugated plate flame arrester under inert conditions [J]. Fuel, 2024, 362: 130822. DOI: 10.1016/j.fuel.2023.130822., articleTitle=null, refAbstract=null), Reference(id=1315732096681890648, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[28], rfOrder=29, authorNames=null, journalName=null, refType=null, unstructuredReference=NIE Z H, GAO W, JIANG H P, et al. Explosion quenching characteristics of corrugated plate microchannels in ethanol-gasoline/air mixtures [J]. International Journal of Thermal Sciences, 2024, 204: 109220. DOI: 10.1016/j.ijthermalsci.2024.109220., articleTitle=null, refAbstract=null), Reference(id=1315732096757388121, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[29], rfOrder=30, authorNames=null, journalName=null, refType=null, unstructuredReference=孙少辰. 波纹管道阻火器系统的阻爆特性研究 [D]. 大连: 大连理工大学, 2018., articleTitle=null, refAbstract=null), Reference(id=1315732097025823578, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[29], rfOrder=31, authorNames=null, journalName=null, refType=null, unstructuredReference=SUN S C. Research of explosion resistant characteristic in in-line crimped-ribbon flame arrester [D]. Dalian: Dalian University of Technology, 2018., articleTitle=null, refAbstract=null), Reference(id=1315732097109709659, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[30], rfOrder=32, authorNames=null, journalName=null, refType=null, unstructuredReference=万阳. 管道内针对氢-空气火焰的多孔板阻火特性研究 [D]. 合肥: 合肥工业大学, 2020. DOI: 10.27101/d.cnki.ghfgu.2020.001840., articleTitle=null, refAbstract=null), Reference(id=1315732097172624220, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[30], rfOrder=33, authorNames=null, journalName=null, refType=null, unstructuredReference=WAN Y. Study on fire resistance characteristics of perforated plate for hydrogen-air flame in a channel [D]. Hefei: Hefei University of Technology, 2020. DOI: 10.27101/d.cnki.ghfgu.2020.001840., articleTitle=null, refAbstract=null), Reference(id=1315732097252315997, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[31], rfOrder=34, authorNames=null, journalName=null, refType=null, unstructuredReference=张炼卓, 闫兴清, 吕先舒, 等. 不同初始压力下可爆预混气体通过波纹板阻火器的淬熄特性研究 [J]. 爆炸与冲击, 2023, 43(1): 012202. DOI: 10.11883/bzycj-2021-4058., articleTitle=null, refAbstract=null), Reference(id=1315732097327813470, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[31], rfOrder=35, authorNames=null, journalName=null, refType=null, unstructuredReference=ZHANG L Z, YAN X Q, LV X S, et al. On quenching characteristics of combustible premixed gas through a crimped-ribbon flame arrester at different initial pressures [J]. Explosion and Shock Waves, 2023, 43(1): 012202. DOI: 10.11883/bzycj-2021-4058., articleTitle=null, refAbstract=null), Reference(id=1315732097386533727, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[32], rfOrder=36, authorNames=null, journalName=null, refType=null, unstructuredReference=ZHONG X T, GALLAGHER B, EVES K, et al. A study of real-world micrograph data quality and machine learning model robustness [J]. NPJ Computational Materials, 2021, 7(1): 161. DOI: 10.1038/s41524-021-00616-3., articleTitle=null, refAbstract=null), Reference(id=1315732097457836896, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[33], rfOrder=37, authorNames=null, journalName=null, refType=null, unstructuredReference=PEI X L, ZHAO Y H, CHEN L W, et al. Robustness of machine learning to color, size change, normalization, and image enhancement on micrograph datasets with large sample differences [J]. Materials & Design, 2023, 232: 112086. DOI: 10.1016/j.matdes.2023.112086., articleTitle=null, refAbstract=null), Reference(id=1315732097529140065, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[34], rfOrder=38, authorNames=null, journalName=null, refType=null, unstructuredReference=VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need [C]//Proceedings of the 31st International Conference on Neural Information Processing Systems. Long Beach: Curran Associates Inc. , 2017: 6000–6010., articleTitle=null, refAbstract=null), Reference(id=1315732097600443234, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[35], rfOrder=39, authorNames=null, journalName=null, refType=null, unstructuredReference=WAIBEL A, HANAZAWA T, HINTON G, et al. Phoneme recognition using time-delay neural networks [J]. IEEE Transactions on Acoustics, Speech, and Signal Processing, 1989, 37(3): 328–339. DOI: 10.1109/29.21701., articleTitle=null, refAbstract=null), Reference(id=1315732097671746403, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[36], rfOrder=40, authorNames=null, journalName=null, refType=null, unstructuredReference=卢祖文. 基于X光图像与深度学习的耐张线夹缺陷识别 [D]. 南昌: 南昌大学, 2022. DOI: 10.27232/d.cnki.gnchu.2022.002699., articleTitle=null, refAbstract=null), Reference(id=1315732097738855268, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[36], rfOrder=41, authorNames=null, journalName=null, refType=null, unstructuredReference=LU Z W. Recognition of tension clamp defects based on X-ray image and deep learning [D]. Nanchang: Nanchang University, 2022. DOI: 10.27232/d.cnki.gnchu.2022.002699., articleTitle=null, refAbstract=null), Reference(id=1315732097810158437, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[37], rfOrder=42, authorNames=null, journalName=null, refType=null, unstructuredReference=BI Y B, WANG Y B, WANG S L, et al. A hybrid expert neural network for predicting hydrogen concentration under the ceiling in underground garage [J]. Building and Environment, 2025, 268: 112325. DOI: 10.1016/j.buildenv.2024.112325., articleTitle=null, refAbstract=null)], funds=null, companyList=[AuthorCompany(id=1315732084048646907, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, xref=1, ext=[AuthorCompanyExt(id=1315732084057035516, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, companyId=1315732084048646907, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1Dalian University of Technology, Dalian 116024, Liaoning, China), AuthorCompanyExt(id=1315732084065424125, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, companyId=1315732084048646907, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1大连理工大学化工学院,辽宁省 大连市 116024)]), AuthorCompany(id=1315732084157698814, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, xref=2, ext=[AuthorCompanyExt(id=1315732084161893119, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, companyId=1315732084157698814, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Beijing Institute of Aerospace Testing Technology, Beijing 100074, China), AuthorCompanyExt(id=1315732084170281728, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, companyId=1315732084157698814, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2北京航天试验技术研究所,北京 100074)])], figs=[ArticleFig(id=1315732087374730014, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=EN, label=Fig.1, caption=Pipeline combustible gas explosion and flame quenching test platform, figureFileSmall=q+7LCawAF6xief1b+87VlA==, figureFileBig=DuSkmUyql8UVf7MyFkKzog==, tableContent=null), ArticleFig(id=1315732087441838879, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=CN, label=图1, caption=管道可燃气燃爆阻火实验平台, figureFileSmall=q+7LCawAF6xief1b+87VlA==, figureFileBig=DuSkmUyql8UVf7MyFkKzog==, tableContent=null), ArticleFig(id=1315732087857074977, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=EN, label=Fig.2, caption=Core architecture of the Transformer model, figureFileSmall=qSL59+SY969PT4SUmVZ0hQ==, figureFileBig=Orr3ub7F/Rpsclnpw3yKxg==, tableContent=null), ArticleFig(id=1315732087957738274, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=CN, label=图2, caption=Transformer模型的核心架构, figureFileSmall=qSL59+SY969PT4SUmVZ0hQ==, figureFileBig=Orr3ub7F/Rpsclnpw3yKxg==, tableContent=null), ArticleFig(id=1315732088221979427, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=EN, label=Fig.3, caption=Core architecture of the CNN model, figureFileSmall=dowXQEYUcj6bZqw8eDTwyA==, figureFileBig=aZLN0MyE3Z9/ewwDDjlQyQ==, tableContent=null), ArticleFig(id=1315732088343614245, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=CN, label=图3, caption=CNN模型的核心架构, figureFileSmall=dowXQEYUcj6bZqw8eDTwyA==, figureFileBig=aZLN0MyE3Z9/ewwDDjlQyQ==, tableContent=null), ArticleFig(id=1315732088674964262, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=EN, label=Fig.4, caption=Impact of different batch_size on Transformer model performance, figureFileSmall=6+VLShLFjNYUCEo42YkKiQ==, figureFileBig=F3FMIMAhiesRwWMX9CZB+g==, tableContent=null), ArticleFig(id=1315732088758850343, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=CN, label=图4, caption=不同批量大小对Transformer模型性能的影响, figureFileSmall=6+VLShLFjNYUCEo42YkKiQ==, figureFileBig=F3FMIMAhiesRwWMX9CZB+g==, tableContent=null), ArticleFig(id=1315732089048257320, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=EN, label=Fig.5, caption=Impact of different epochs on the performances of Transformer models, figureFileSmall=W+iToM4eKDGniZk+uuLA5g==, figureFileBig=xNMC2taw60mwI7PeErFT+Q==, tableContent=null), ArticleFig(id=1315732089140532009, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=CN, label=图5, caption=不同训练轮次对Transformer模型性能的影响, figureFileSmall=W+iToM4eKDGniZk+uuLA5g==, figureFileBig=xNMC2taw60mwI7PeErFT+Q==, tableContent=null), ArticleFig(id=1315732089216029482, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=EN, label=Fig.6, caption=Comparison of Transformer model predictions and actual values, figureFileSmall=cvJswzWpCkNHyabQhB00lQ==, figureFileBig=FOzcENYBnndTAAwV4onqmw==, tableContent=null), ArticleFig(id=1315732089283138347, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=CN, label=图6, caption=Transformer模型预测值与真实值比较, figureFileSmall=cvJswzWpCkNHyabQhB00lQ==, figureFileBig=FOzcENYBnndTAAwV4onqmw==, tableContent=null), ArticleFig(id=1315732089530602284, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=EN, label=Fig.7, caption=Comparison of CNN model predictions and actual values, figureFileSmall=Id2PN1dQguLdro6oEdoyaQ==, figureFileBig=ZDeztutkzJaMGSWDmzO1mA==, tableContent=null), ArticleFig(id=1315732089606099757, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=CN, label=图7, caption=CNN模型预测值与真实值比较, figureFileSmall=Id2PN1dQguLdro6oEdoyaQ==, figureFileBig=ZDeztutkzJaMGSWDmzO1mA==, tableContent=null), ArticleFig(id=1315732089694180142, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=EN, label=Fig.8, caption=Distribution of evaluation indicators for the CNN and Transformer models, figureFileSmall=n03zY56J5BGl+kmBygt1Tg==, figureFileBig=O+JxAJ89eXp4h3pThKaxpg==, tableContent=null), ArticleFig(id=1315732089887118127, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=CN, label=图8, caption=两种模型的评估指标分布, figureFileSmall=n03zY56J5BGl+kmBygt1Tg==, figureFileBig=O+JxAJ89eXp4h3pThKaxpg==, tableContent=null), ArticleFig(id=1315732089962615600, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=EN, label=Fig.9, caption=Comparison of experimental values and error values between the CNN and Transformer models, figureFileSmall=AFUsuRdcHFngrwch9eFZBw==, figureFileBig=kUzmhkl2orHISev9qR4MRg==, tableContent=null), ArticleFig(id=1315732090046501681, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=CN, label=图9, caption=2种模型的实验值和误差值比较, figureFileSmall=AFUsuRdcHFngrwch9eFZBw==, figureFileBig=kUzmhkl2orHISev9qR4MRg==, tableContent=null), ArticleFig(id=1315732091715834674, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=EN, label=Fig.10, caption=Impact of normalization on the performance of the CNN and Transformer models, figureFileSmall=rSPIJCmwcKAJqk8O26yehg==, figureFileBig=ZGbgTn77T8z2Z2MchT1Plg==, tableContent=null), ArticleFig(id=1315732091980075827, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=CN, label=图10, caption=归一化处理对两种模型性能的影响, figureFileSmall=rSPIJCmwcKAJqk8O26yehg==, figureFileBig=ZGbgTn77T8z2Z2MchT1Plg==, tableContent=null), ArticleFig(id=1315732092055573300, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=EN, label=Tab.1, caption=

Abstract of gas critical quenching diameter experiment database

, figureFileSmall=null, figureFileBig=null, tableContent=
来源燃气体积分数/%长径比初始压力/kPa多孔介质厚度/mm热导率/(W·m−1·K−1)
文献[27]14.6~38.714.188101.32540~12016.2
文献[30]17.4~42.47.14320~10020~100238
文献[29]27.430~40101.32530~12016.2
文献[29]4.250~70101.32550~15016.2
文献[28]1.2~2.514.188101.32540~12016.2
文献[31]5.0~9.55030~603816.2
文献[31]4.25070~803816.2
), ArticleFig(id=1315732092126876469, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=CN, label=表1, caption=

燃气临界淬熄直径实验数据库摘要

, figureFileSmall=null, figureFileBig=null, tableContent=
来源燃气体积分数/%长径比初始压力/kPa多孔介质厚度/mm热导率/(W·m−1·K−1)
文献[27]14.6~38.714.188101.32540~12016.2
文献[30]17.4~42.47.14320~10020~100238
文献[29]27.430~40101.32530~12016.2
文献[29]4.250~70101.32550~15016.2
文献[28]1.2~2.514.188101.32540~12016.2
文献[31]5.0~9.55030~603816.2
文献[31]4.25070~803816.2
), ArticleFig(id=1315732092198179638, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=EN, label=Tab.2, caption=

Optimal hyperparameter combinations for Transformer models

, figureFileSmall=null, figureFileBig=null, tableContent=
超参数最优值
批量大小32
训练轮次500
), ArticleFig(id=1315732092277871415, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=CN, label=表2, caption=

Transformer模型最优超参数组合

, figureFileSmall=null, figureFileBig=null, tableContent=
超参数最优值
批量大小32
训练轮次500
), ArticleFig(id=1315732092445643576, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=EN, label=Tab.3, caption=

Optimal hyperparameter combination for CNN models

, figureFileSmall=null, figureFileBig=null, tableContent=
超参数最优值
批量大小16
训练轮次600
), ArticleFig(id=1315732092504363833, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, language=CN, label=表3, caption=

CNN模型最优超参数组合

, figureFileSmall=null, figureFileBig=null, tableContent=
超参数最优值
批量大小16
训练轮次600
)], attaches=null, journal=Journal(id=1315121464548098176, delFlag=0, nameCn=爆炸与冲击, nameEn=Explosion and Shock Waves, nameHistory1=null, nameHistory2=null, issn=1001-1455, eissn=null, cn=51-1148/O3, coden=null, periodic=0, language=CN, oaType=null, ccby=null, superviseOffice=null, ownerOffice=null, pubOffice=null, editorOffice=null, officeType=null, aims=null, clcCode=null, officeProv=null, officeCity=null, officeAddr=null, officeZip=null, officeEmail=null, officePhone=null, editDirector=null, officeDirector=null, officeDirectorPhone=null, officeStaffNum=null, officeEmpNum=null, coverPicUrl=7AR1jUjcsRcIbhldpwIErw==, journalPrice=null, startedYear=null, abbrevIsoEn=Explosion and Shock Waves, journalRemark=null, publicationField=null, createdTime=1791478990925, updatedTime=1791480098923, createdBy=18614031015, updatedBy=13701087609, firstLetterCn=B, firstLetterEn=B, subjectCode=Engineering, subjectName=null, subjectCodeEn=Engineering, subjectNameEn=null, picCn=7AR1jUjcsRcIbhldpwIErw==, picEn=mzg+w7cy0OwnBDUP5+hcoQ==, jcr=null, cjcr=null, exts=[JournalExt(id=1315126111902745388, language=CN, name=爆炸与冲击, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1791480098939, updatedTime=1791480098939, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://bzycjauthor.cast.org.cn/webm, submissionEditorUrl=https://bzycjeditor.cast.org.cn/webm, submissionReviewUrl=https://bzycjauthor.cast.org.cn/webm, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1315126111961465645, language=EN, name=Explosion and Shock Waves, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1791480098953, updatedTime=1791480098953, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://bzycjauthor.cast.org.cn/webm, submissionEditorUrl=https://bzycjeditor.cast.org.cn/webm, submissionReviewUrl=https://bzycjauthor.cast.org.cn/webm, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1315122079294652549, websiteList=[Website(id=1315346231975596681, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1315122079294652549, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/bzycj/CN, language=CN, createTime=1791532579654, createBy=18614031015, updateTime=1791533562260, updateBy=18614031015, name=爆炸与冲击-中文, tplId=1146099689490845704, title=爆炸与冲击, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1315350498421617264, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1315346231975596681, code=articleTextType, value=kx, createTime=1791533596854, updateTime=1791533596854, creator=18614031015, updator=18614031015), WebsiteProps(id=1315350498392257133, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1315346231975596681, code=banner, value=null, createTime=1791533596847, updateTime=1791533596847, creator=18614031015, updator=18614031015), WebsiteProps(id=1315350498446783091, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1315346231975596681, code=grayFlag, value=0, createTime=1791533596860, updateTime=1791533596860, creator=18614031015, updator=18614031015), WebsiteProps(id=1315350498383868524, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1315346231975596681, code=logo, value=https://castjournals.cast.org.cn/joweb/bzycj/CN/file/pic?fileId=FJYiT+NKwp65EbTOf64bOg==, createTime=1791533596845, updateTime=1791533596845, creator=18614031015, updator=18614031015), WebsiteProps(id=1315350498463560309, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1315346231975596681, code=minRunFlag, value=0, createTime=1791533596864, updateTime=1791533596864, creator=18614031015, updator=18614031015), WebsiteProps(id=1315350498413228655, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1315346231975596681, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/bzycj/CN/file/pic, createTime=1791533596852, updateTime=1791533596852, creator=18614031015, updator=18614031015), WebsiteProps(id=1315350498455171700, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1315346231975596681, code=silenceFlag, value=0, createTime=1791533596862, updateTime=1791533596862, creator=18614031015, updator=18614031015), WebsiteProps(id=1315350498400645742, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1315346231975596681, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1791533596849, updateTime=1791533596849, creator=18614031015, updator=18614031015), WebsiteProps(id=1315350498430005873, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1315346231975596681, code=themeColor, value=null, createTime=1791533596856, updateTime=1791533596856, creator=18614031015, updator=18614031015), WebsiteProps(id=1315350498438394482, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1315346231975596681, code=themeStyle, value=null, createTime=1791533596858, updateTime=1791533596858, creator=18614031015, updator=18614031015)]), Website(id=1315346232025928330, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1315122079294652549, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/bzycj/EN, language=EN, createTime=1791532579666, createBy=18614031015, updateTime=1791533558083, updateBy=18614031015, name=爆炸与冲击-英文, tplId=1146101810881728533, title=Explosion and Shock Waves, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1315350548728099453, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1315346232025928330, code=articleTextType, value=kx, createTime=1791533608848, updateTime=1791533608848, creator=18614031015, updator=18614031015), WebsiteProps(id=1315350548707127930, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1315346232025928330, code=banner, value=null, createTime=1791533608843, updateTime=1791533608843, creator=18614031015, updator=18614031015), WebsiteProps(id=1315350548757459584, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1315346232025928330, code=grayFlag, value=0, createTime=1791533608855, updateTime=1791533608855, creator=18614031015, updator=18614031015), WebsiteProps(id=1315350548698739321, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1315346232025928330, code=logo, value=https://castjournals.cast.org.cn/joweb/bzycj/EN/file/pic?fileId=FJYiT+NKwp65EbTOf64bOg==, createTime=1791533608841, updateTime=1791533608841, creator=18614031015, updator=18614031015), WebsiteProps(id=1315350548807791234, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1315346232025928330, code=minRunFlag, value=0, createTime=1791533608867, updateTime=1791533608867, creator=18614031015, updator=18614031015), WebsiteProps(id=1315350548719710844, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1315346232025928330, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/bzycj/EN/file/pic, createTime=1791533608846, updateTime=1791533608846, creator=18614031015, updator=18614031015), WebsiteProps(id=1315350548791014017, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1315346232025928330, code=silenceFlag, value=0, createTime=1791533608863, updateTime=1791533608863, creator=18614031015, updator=18614031015), WebsiteProps(id=1315350548715516539, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1315346232025928330, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1791533608845, updateTime=1791533608845, creator=18614031015, updator=18614031015), WebsiteProps(id=1315350548736488062, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1315346232025928330, code=themeColor, value=null, createTime=1791533608850, updateTime=1791533608850, creator=18614031015, updator=18614031015), WebsiteProps(id=1315350548744876671, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1315346232025928330, code=themeStyle, value=null, createTime=1791533608852, updateTime=1791533608852, creator=18614031015, updator=18614031015)])], journalTitle=爆炸与冲击, weixinUrl=null, journalUrl=https://bzycj.cn/, iacademicId=null, status=1, seqNo=null, journalTitleEn=Explosion and Shock Waves, journalPhotoCn=7AR1jUjcsRcIbhldpwIErw==, journalPhotoEn=mzg+w7cy0OwnBDUP5+hcoQ==, journalFirstLetter=B, journalRecommend=null, journalNew=null, journalCollection=null, jcrJf=null, cjcrJf=null, jcrJfStr=null, cjcrJfStr=null, submissionFirstDecision=null, sciSubjectClassification=null, casSubjectClassification=null, citeScore=null, totalCitationFrequency=null, icpCode=null, psCode=null, advertisingLicenseCode=null, copyrightInformation=null, country=null, option=, provinceCode=null, provinceName=null, collectFlag=false, interPubPlatform=, interPubPlatformUrl=null), detailUrlCn=https://castjournals.cast.org.cn/joweb/bzycj/CN/10.11883/bzycj-2025-0218, detailUrlEn=https://castjournals.cast.org.cn/joweb/bzycj/EN/10.11883/bzycj-2025-0218, pdfUrlCn=https://castjournals.cast.org.cn/joweb/bzycj/CN/PDF/10.11883/bzycj-2025-0218, pdfUrlEn=https://castjournals.cast.org.cn/joweb/bzycj/EN/PDF/10.11883/bzycj-2025-0218, aliStartDate=0, aliEndDate=0, collectionFlag=false, citedCount=null, citedUrl=null, previewStatus=0, delFlag=0, hasFullText=1, orderTime=1788537600000, fullTextJson=null, articleText=null, reference=null)
收藏切换
基于深度学习算法预测燃爆的临界淬熄直径
收藏切换
PDF下载
聂仲恒 1, 2 , 汪丽 2 , 高伟 1 , 姜海鹏 1
爆炸与冲击 | 学科交叉前沿 2026,46(9): 091502-1-091502-12
收起
收藏切换
爆炸与冲击 |学科交叉前沿 2026 , 46 (9) : 091502-1 -091502-12
基于深度学习算法预测燃爆的临界淬熄直径
全屏
[Author(id=1315732084233196290, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=nzh2490823579@163.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1315732084300305157, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, authorId=1315732084233196290, language=EN, stringName=Zhongheng NIE, firstName=Zhongheng, middleName=null, lastName=NIE, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1Dalian University of Technology, Dalian 116024, Liaoning, China
2Beijing Institute of Aerospace Testing Technology, Beijing 100074, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1315732084371608326, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, authorId=1315732084233196290, language=CN, stringName=聂仲恒, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1大连理工大学化工学院,辽宁省 大连市 116024
2北京航天试验技术研究所,北京 100074, bio={"content":"

聂仲恒(2000- ),男,硕士,

"}, bioImg=null, bioContent=

聂仲恒(2000- ),男,硕士,

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1315732084048646907, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, xref=1, ext=[AuthorCompanyExt(id=1315732084057035516, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, companyId=1315732084048646907, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1Dalian University of Technology, Dalian 116024, Liaoning, China), AuthorCompanyExt(id=1315732084065424125, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, companyId=1315732084048646907, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1大连理工大学化工学院,辽宁省 大连市 116024)]), AuthorCompany(id=1315732084157698814, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, xref=2, ext=[AuthorCompanyExt(id=1315732084161893119, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, companyId=1315732084157698814, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Beijing Institute of Aerospace Testing Technology, Beijing 100074, China), AuthorCompanyExt(id=1315732084170281728, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, companyId=1315732084157698814, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2北京航天试验技术研究所,北京 100074)])]), Author(id=1315732084438717192, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1315732084510020362, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, authorId=1315732084438717192, language=EN, stringName=Li WANG, firstName=Li, middleName=null, lastName=WANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2Beijing Institute of Aerospace Testing Technology, Beijing 100074, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1315732084585517835, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, authorId=1315732084438717192, language=CN, stringName=汪丽, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2北京航天试验技术研究所,北京 100074, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1315732084157698814, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, xref=2, ext=[AuthorCompanyExt(id=1315732084161893119, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, companyId=1315732084157698814, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Beijing Institute of Aerospace Testing Technology, Beijing 100074, China), AuthorCompanyExt(id=1315732084170281728, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, companyId=1315732084157698814, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2北京航天试验技术研究所,北京 100074)])]), Author(id=1315732084652626701, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1315732084728124175, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, authorId=1315732084652626701, language=EN, stringName=Wei GAO, firstName=Wei, middleName=null, lastName=GAO, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1Dalian University of Technology, Dalian 116024, Liaoning, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1315732084803621648, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, authorId=1315732084652626701, language=CN, stringName=高伟, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1大连理工大学化工学院,辽宁省 大连市 116024, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1315732084048646907, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, xref=1, ext=[AuthorCompanyExt(id=1315732084057035516, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, companyId=1315732084048646907, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1Dalian University of Technology, Dalian 116024, Liaoning, China), AuthorCompanyExt(id=1315732084065424125, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, companyId=1315732084048646907, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1大连理工大学化工学院,辽宁省 大连市 116024)])]), Author(id=1315732084883313426, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=jhp@dlut.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1315732084963005204, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, authorId=1315732084883313426, language=EN, stringName=Haipeng JIANG, firstName=Haipeng, middleName=null, lastName=JIANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1Dalian University of Technology, Dalian 116024, Liaoning, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1315732085030114069, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, authorId=1315732084883313426, language=CN, stringName=姜海鹏, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1大连理工大学化工学院,辽宁省 大连市 116024, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1315732084048646907, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, xref=1, ext=[AuthorCompanyExt(id=1315732084057035516, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, companyId=1315732084048646907, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1Dalian University of Technology, Dalian 116024, Liaoning, China), AuthorCompanyExt(id=1315732084065424125, tenantId=1146029695717560320, journalId=1315122079294652549, articleId=1315732079397163745, companyId=1315732084048646907, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1大连理工大学化工学院,辽宁省 大连市 116024)])])]
聂仲恒1, 2 , 汪丽2, 高伟1, 姜海鹏1
作者信息
  • 1大连理工大学化工学院,辽宁省 大连市 116024
  • 2北京航天试验技术研究所,北京 100074
通讯作者:
姜海鹏(1990- ),男,博士,副教授,
作者简介:

聂仲恒(2000- ),男,硕士,

Prediction of critical quenching diameter for deflagration and detonation based on deep learning algorithms
Zhongheng NIE1, 2 , Li WANG2, Wei GAO1, Haipeng JIANG1
Affiliations
  • 1Dalian University of Technology, Dalian 116024, Liaoning, China
  • 2Beijing Institute of Aerospace Testing Technology, Beijing 100074, China
出版时间: 2026-09-05 doi: 10.11883/bzycj-2025-0218
文章导航
收藏切换

聚焦密闭管道可燃气体燃爆火焰淬熄的安全防护需求,基于燃气组分、管道几何、初始条件及多孔介质结构参数构建了九维特征空间,建立了多孔介质临界淬熄直径预测模型。为突破传统经验公式的精度不足问题与应用局限性,通过系统性的超参数优化与模型验证,深入对比并证实了Transformer架构在临界淬熄直径预测问题上的显著优越性:其预测性能(平均绝对误差δMAE=0.068,均方误差δMSE=0.008,相关系数R2=0.928)显著超越了广泛应用的卷积神经网络(convolutional neural network,CNN)模型(δMAE=0.079,δMSE=0.012,R2=0.906),不仅整体误差更低,且对离群点的鲁棒性更强。深入分析发现,Transformer模型的核心优势源于其自注意力机制对淬熄过程中关键临界特征的精准捕获与高效建模能力。在数据归一化敏感性验证中,Transformer模型展现出优异的鲁棒性,这归功于其层归一化机制所赋予的特征解耦与稳定表示能力。基于上述系统性评估,最终确立Transformer模型为预测多孔介质临界淬熄直径的最优模型,可为燃爆安全防控策略的量化制定及管道阻火器安全性能的精细化设计提供强大的、可操作的决策支持工具,具有重要的理论指导意义。

管道爆炸  /  深度学习  /  临界淬熄直径预测  /  阻火器

The research addresses the safety imperative of preventing flammable gas explosions in enclosed pipelines by establishing a predictive model for the critical quenching diameter within porous media flame arresters. A novel predictive framework was developed based on a comprehensive nine-dimensional feature space incorporating gas composition parameters (e.g., hydrogen equivalence ratio), pipeline geometry dimensions (length-to-diameter ratio), initial thermodynamic conditions (pressure), and porous medium structural characteristics (thickness, material thermal conductivity). A systematic investigation was conducted to identify the optimal hyperparameter configurations for both convolutional neural network (CNN) and Transformer architectures. Rigorous validation demonstrated that the Transformer model exhibits statistically significant superiority over the CNN model across all key performance metrics. Specifically, the model achieved a δMAE (mean absolute error, MAE) of 0.068, a δMSE (mean squared error, MSE) of 0.008, and an coefficient of determination R2 of 0.928. The performance notably surpassed the CNN results (δMAE=0.079, δMSE=0.012, R2=0.906). Beyond evaluation indicators, detailed error distribution analysis confirmed the Transformer's enhanced predictive accuracy and reduced susceptibility to outliers. The superior performance is attributed to the Transformer’s intrinsic self-attention mechanism, which excels at dynamically identifying and weighting critical interdependencies among the diverse input features governing the complex quenching process. The capability enables more precise capture of the nonlinear phenomena defining the quenching limit. Furthermore, robustness testing involving diverse data normalization schemes revealed that the Transformer model exhibits greater stability. This resilience stems from its inherent layer normalization mechanism, which effectively decouples feature dependencies and mitigates sensitivity to input scaling variations. Consequently, the Transformer architecture is established as the definitive optimal model for this critical safety prediction task. Its significant advantage lies in the minimal data preprocessing required before deployment, a feature that greatly enhances its practical utility. The model provides robust quantitative decision-making support essential for formulating effective gas explosion mitigation strategies and optimizing the safety design parameters of pipeline flame arresters. By accurately predicting the critical quenching diameter across various scenarios, this work delivers a valuable tool for enhancing inherent safety in industries handling combustible gases within confined pipeline systems.

pipeline explosion  /  deep learning  /  critical quenching diameter prediction  /  flame arrester
聂仲恒, 汪丽, 高伟, 姜海鹏. 基于深度学习算法预测燃爆的临界淬熄直径. 爆炸与冲击, 2026 , 46 (9) : 091502-1 -091502-12 . DOI: 10.11883/bzycj-2025-0218
Zhongheng NIE, Li WANG, Wei GAO, Haipeng JIANG. Prediction of critical quenching diameter for deflagration and detonation based on deep learning algorithms[J]. Explosion and Shock Waves, 2026 , 46 (9) : 091502-1 -091502-12 . DOI: 10.11883/bzycj-2025-0218
对于降低管道内可燃气的爆炸风险,管道阻火器的安装与应用至关重要。阻火器能有效阻止火焰沿管道传播,从而提升安全性[1]。然而,不同可燃气体的理化性质存在显著差异,这为阻火器的设计带来了挑战[2]。同时,各类阻火介质(如波纹板、泡沫金属、多孔板等)的阻火性能亦呈现显著差别[3-5]。由此可见,可燃气体种类与浓度、阻火介质类型、厚度及孔隙率等因素构成了阻火器设计过程中的核心难点[6-7]。然而,现有研究在定量确定多孔介质的临界淬熄直径方面仍面临困难。所幸,计算机技术的发展与各类深度学习算法的兴起,为解决此问题提供了可能[8-11]。
在当前研究阶段,火焰淬熄机理主要依托于传热理论以及化学链式反应理论这两大理论体系[12]。依据传热理论所确立的淬熄判定标准[13-14]为:爆炸过程所产生的放热量应当小于或等于火焰与固体之间的传热量。然而,就实际应用情境而言,使用者往往更关切阻火介质所展现出的阻火性能。在此现实需求的驱动之下,学者们纷纷投身于探究阻火介质种类、阻火介质厚度以及孔隙率等关键因素对于淬熄结果所产生的影响[15-16]。经由一系列严谨的实验研究,众多研究者相继提出了一些具有关键意义的临界淬熄参数,诸如安全阻火速度[17-18]、临界淬熄直径[19]和临界淬熄厚度[20-21]等经验公式。例如:在对阻火器临界淬熄直径的深入探究过程中,研究者们通过大量的实验观测发现,淬熄直径与湍流火焰速度呈现出负相关关系,并据此推导出了相应的经验公式[22]。不容忽视的是,这些经验公式在工业安全防护中均存在着局限性。鉴于此,从纯粹的理论计算视角出发、精准地获取适用于不同种类可燃气体以及不同淬熄介质条件下的临界淬熄直径,尚存在巨大的困难。
随着机器学习的高速发展,目前机器学习已被积极应用于爆炸安全防护的各个方面[23-25]。Zhang等[10]建立了用于预测甲烷爆炸压力的KPCA-GWO-LSSVM模型。Xu等[26]结合理论分析与机器学习构建了F-P-PSO-DNN模型,建立了气体类型与浓度、容器参数及点火位置与泄爆超压的关系,该模型的准确性仅比莫尔科夫模型低19%。因此,从算法原理的角度来看,机器学习算法有利于多因素条件下的临界淬熄直径预测。然而,目前关于多孔介质中临界淬熄直径预测的相关研究较少,有待进一步研究。
基于上述研究背景,本文致力于构建Transformer架构与卷积神经网络(convolutional neural network,CNN)模型,开展可燃气体临界淬熄直径的精确预测工作。通过系统评估2种回归模型的预测效能,判断适配于临界淬熄直径预测的深度学习模型。该模型以可燃气类别、体积分数、初始压力、阻火介质厚度以及惰性气体注入状况等多维度变量为输入依据,预测不同可燃气体系及阻火介质结构参数组合下的临界淬熄直径,以期为可燃气体燃爆安全防控策略的制定以及管道阻火器安全设计的优化提供量化决策支持。
数据来源于2部分:自建实验平台获取的数据及文献数据,其中自建实验数据占主体。为了开展管道可燃气体阻火特性研究,搭建了如图1所示的密闭管道可燃气阻火实验平台,该平台具备密闭管道内可燃气体点火、爆燃火焰传播特性记录及压力动态监测功能:实验管道长径比为14.188;压力传感器布设于阻火单元前后,用于捕捉爆燃压力演化过程;火焰信号则由光电传感器采集。阻火效能判定以编号F.4的光电传感器信号特征消失为关键依据。由于氢气和乙醇汽油的物化性质不同,使用燃气浓度和密度区分2种可燃气。基于此平台,系统获取了初始压力为常压、阻火介质为波纹板(厚度范围为40~120 mm)、不同浓度可燃气爆燃条件下的阻火实验数据集[27-28]。
补充实验数据来源于现有的文献数据库,源数据量为162个,如表1所示。其中,孙少辰[29]考察了波纹板厚度与孔隙率对燃爆火焰传播特性的影响,其实验参数为:管道长径比为30~40,初始压力常压,波纹板厚度范围为30~120 mm。需特别说明的是,研究中氢气浓度为固定值。此外,参考万阳[30]的实验数据,其差异点在于:阻火介质为多孔板,初始压力为20~100 kPa的增压环境,多孔板厚度范围为20~100 mm。为明确区分波纹板与多孔板2类阻火介质,引入其导热率作为关键表征参数。此外,还补充了丙烷和乙烯的阻火数据[31],其多孔介质厚度均为38 mm,以改变初始压力实现火焰淬熄。值得一提的是,由于波纹板与多孔板的孔隙率与微孔道直径成正比,因此,预测目标以临界孔隙率指代。综上,聚焦于探究燃气种类(氢气、乙醇汽油、丙烷和乙烯)与浓度、管道长径比、初始压力、多孔介质厚度及其导热率对可燃气临界淬熄直径的影响规律。
对于多孔介质的临界淬熄直径预测,通过实验获取的初始数据存在显著的离散性(如多孔介质厚度波动范围达15~85 mm),其原因是实验装置尺寸差异、燃气种类变化(氢气、乙醇汽油、丙烷和乙烯)及初始条件多样性(压力为20~100 kPa)。为提升深度学习模型的鲁棒性,首先剔除异常值,消除极端工况对统计规律的干扰。最终确立九维输入特征体系:燃气体积分数、燃气密度、管径、长径比、初始压力、氮气体积分数、二氧化碳体积分数、多孔介质厚度及热导率,输出目标为临界淬熄直径。其中,初始压力为每次实验开始时管道的初始压力,会对淬熄结果有重大影响;多孔介质厚度和热导率是衡量多孔介质淬熄能力的重要参数。为了预测燃气临界淬熄直径,对9个输入特征与输出目标(临界淬熄直径)进行最小-最大归一化预处理[32],其数学原理为:
$ {\ddot{X}}_{i}=\frac{{X}_{i}-{X}_{\min }}{{X}_{\max }-{X}_{\min }} $
式中:$ {\ddot{X}} $为归一化数据,Xi为预处理前的数据,Xmin和Xmax分别为特征数据中的最小值和最大值。该方法通过线性变换将各特征压缩至[0,1]区间,其核心作用在于消除多量纲特征尺度的差异(如管径与热导率的数值量级悬殊),优化梯度下降收敛效率,并提升模型对特征贡献度的均衡学习能力,从而增强预测鲁棒性,为深度学习模型构建奠定数据基础[33]。
Transformer模型是2017年由Vaswani等[34]提出的神经网络架构,其核心结构如图2所示。该模型首次提出依赖自注意力(self-attention)机制来建模数据元素间的全局依赖关系。核心思想是处理输入数据中每个元素相对于所有其他元素的注意力权重,从而动态地聚合上下文信息并生成具有丰富关联性的表示。自注意力过程的核心计算为:
$ {Attention}(\boldsymbol{Q},\boldsymbol{K},\boldsymbol{V})={soft}\max \left(\frac{\boldsymbol{Q}\boldsymbol{K}^\text{T}}{\sqrt{{d}_{k}}}\right){\boldsymbol{V}} $
式中:查询Q、键K、值V向量通过线性变换从输入嵌入获得,注意力得分通过缩放点积计算并归一化。此外,Transformer模型通过位置编码来显式注入序列的顺序信息,弥补了自注意力机制本身对位置不敏感的缺陷。对于结构化表格数据或特征工程后的序列,可以构建一个简化的Transformer模型变体。这种变体通常利用多层感知机(multi-layer perceptron,MLP)或线性层模拟注意力机制对特征交互的捕获能力,并结合残差连接与层归一化来稳定深层训练。模型的深度、隐藏层维度以及正则化策略(如Dropout、权重衰减)需根据具体预测任务的复杂度和数据特性进行细致调整。关键前提是确保输入特征经过适当的预处理(如标准化),且其维度与模型输入层的要求严格兼容。
CNN模型是受生物视觉皮层层次化感知机制启发而设计的深度学习模型[35]。标准CNN模型由特征提取器与分类器构成,核心包含3类层级:卷积层、池化层和全连接层[36],核心架构如图3所示。从输入层到输出层进行层级化特征抽象,卷积层通过局部感受实现对输入数据的空间特征提取。CNN的训练过程通常采用反向传播算法进行端到端优化,通过卷积核的权值共享机制显著降低模型复杂度,克服了全连接网络在处理高维图像数据时的参数爆炸问题。CNN可与多种非线性激活函数配合使用,选择ReLU激活函数作为卷积层激活函数,能有效抑制梯度消失并提升稀疏表达能力,其数学表达式如下:
$ f\left(x\right)=\max \left(0,x\right) $
所有层均应用L2正则化防止过拟合,其中L2正则化的表达形式为:
$ {L}_\text{total}={L}_\text{data}+\lambda \sum \limits_{i}w_{i}^{2} $
式中:原始损失函数用Ltotal表示,λ为正则化系数。
基于Python语言中的TensorFlow架构,采用归一化九维特征数据集构建卷积神经网络CNN与Transformer模型预测框架。CNN通过一维卷积核与全局平均池化层提取管径-厚度-孔隙率等局部几何约束特征,其离散卷积运算有效捕捉空间尺度关联;Transformer模型则依托自注意力机制,解析燃气与惰性气体体积分数交互的全局非线性耦合效应,突破传统模型对长程依赖建模的局限。超参数优化采用手动调参方法,结合L2正则化抑制过拟合,最终以Adam优化器驱动模型收敛。
在构建燃气临界淬熄直径预测模型后,需对其性能进行系统化评估。采用相关系数R2、平均绝对误差(mean absolute error, MAE)δMAE和均方误差(mean squared error, MSE)δMSE构成综合评价体系[37]。
R2可揭示模型对响应变量变异性的解释能力,其计算基于总离差平方和(sums of squared deviations,SST)与残差平方和(sum of squares due to error,SSE)的归一化比值:
$ {R}^{2}=1-\frac{\displaystyle\sum \limits_{i=1}^{n}{\left({y}_{i}-y_{i}^{\prime}\right)}^{2}}{\displaystyle\sum \limits_{i=1}^{n}{\left({y}_{i}-\overline{y}\right)}^{2}} $
δMAE表征预测值与真实值间绝对偏离程度的期望估计,其定义为:
$ \mathit{\delta}_{{\mathrm{MAE}}}=\frac{1}{n}\sum_{i=1}^n\left|y_i-y_i^{\prime}\right| $
δMSE量化预测误差的二次矩估计,数学表述为:
$ \mathit{\delta}\mathrm{_{MSE}}=\frac{1}{n}\sum \limits_{i=1}^{n}{\left({y}_{i}-y_{i}^{\prime}\right)}^{2} $
式中:yi为实验值,$ y_{i}^{\prime} $为模型预测值,$ \overline{y} $为实验值的平均值,n为实验数据量。
其中,δMAE因采用绝对值运算对异常值具有鲁棒性,单独使用可能弱化模型缺陷的显性表征;而δMSE通过平方运算放大显著偏差的权重。为避免单一指标的评估局限性,综合运用δMAE与δMSE解析临界淬熄直径预测值与实验值的离散特性。R2则用于量化模型对数据变异性的解释能力,其趋近于1表征拟合优度提升。3项指标协同实现了模型泛化能力的多维度验证。
构建基于Transformer的回归预测模型,用于预测多孔介质临界孔隙率,其核心逻辑如下:从标准化数据集中提取9个关键特征参数(包括燃气浓度、管径、初始压力等)作为输入变量,以多孔介质临界孔隙率为目标变量。首先对数据进行标准化预处理,然后按9∶1比例将处理后的数据划分为训练集和测试集。模型架构采用Transformer核心结构:输入层接收特征向量后,依次经过2个全连接隐藏层(512和256个神经元,ReLU激活函数),每层均引入L2正则化(λ=0.005)和Dropout(10%)机制以增强泛化能力,最终通过线性输出层生成预测值。采用Adam优化器进行N个训练轮次(Epoch)的训练(批量大小自定)。模型评估阶段计算测试集的δMAE、δMSE和R2性能指标,最后将训练历史数据、评估结果及完整模型参数分别保存,实现实验过程的可重复性与结果可追溯性。
对于Transformer模型,批量大小和训练轮次的选择会影响模型预测的效率和准确率,因此,需要获得最优的参数组合。通过手动调参,本文获得了最优超参数组合,如表2所示。其中,最优的批量大小和训练轮次分别为32和500。通过固定其他参数,获得了不同超参数对Transformer模型性能的影响如图4和图5所示。
在模型训练过程中,批量大小(每次迭代时同时输入的样本数量)为关键超参数,其取值的确定需综合考量硬件限制条件以及随机性引入程度等因素,以保障模型训练的稳定性。通常情况下,较大的批量大小借助硬件加速优势,可提升训练速度。图4呈现了当批量大小分别设置为16、32、48、64和80时,模型评估参数相应的变动情形。经观察发现,批量大小为32时,δMAE与δMSE均达到最小值,而R2取得最大值。当批量大小由16递增至32,相关性系数呈现逐步上升趋势,增幅较大。然而,当批量大小增至48时,R2急剧回落至0.889,相较于批量大小为32时的R2值,降幅达4.2%,表明过大的批量会削弱模型预测的精准度与泛化能力。但从整体来看,批量大小对Transformer模型性能的影响力有限。权衡各方面因素后,确定最优批量大小为32。
Epoch表征模型遍历完整训练集的次数。通常,模型预测性能随轮次增加而提升,但伴随时间成本上升及过拟合风险增大,进而削弱泛化能力。因此,确定最优轮次至关重要。如图5所示,对比分析400、500、600、700及800五种训练轮次下的性能表现发现,各评估指标随轮次变化幅度较小,表明轮次对整体性能影响有限。在Epoch=500轮次时,模型性能达到最优(δMAE=0.067, δMSE=0.008, R2=0.928)。综合考虑训练效率与泛化能力,最终选定Epoch为500。
获得最优超参数后,将最优的Transformer模型得到的预测值与实验得到的真实值作比较,如图6所示。可以发现,预测值的误差范围大部分在7%以内,存在小部分离群值。这种精度程度表明该模型能够准确地预测多孔介质临界淬熄直径,从而确保应用场景中对管道燃气爆炸的有效阻隔。同时,较低的误差范围也说明了该模型在预测临界淬熄直径方面的稳健性,有利于可燃气体燃爆安全防控策略的制定以及管道阻火器安全设计的优化。
除了Transformer模型,基于一维卷积神经网络(1D CNN)建立回归预测模型,其逻辑流程是:读取标准化后的数据集,选择9个输入特征和多孔介质临界淬熄直径作为预测目标;对特征数据进行标准化处理,并按9∶1的比例划分训练集和测试集;构建包含2个Conv1D层(分别使用ReLU和Sigmoid激活函数)、展平层、256节点全连接层(带LeakyReLU激活)和单节点输出层的网络架构,所有层均应用L2正则化防止过拟合;使用Adam优化器编译模型,在特定Epoch的小批量训练;训练完成后评估测试集性能,计算δMAE、δMSE和R2指标。经过超参数优化得到最优的训练模型。通过调整超参数,本文获得了CNN模型的最优超参数组合,如表3所示。
在确定卷积神经网络CNN模型的最优超参数组合后,对其预测性能进行了严谨的定量验证。如图7所示,模型预测值与真实值对比分析的结果表明,该CNN模型的预测误差总体上控制在约10%的范围内。然而,值得注意的是,预测结果中存在一定数量的离群值。尽管这一精度水平在一定程度上反映了模型的有效性,但对于管道燃爆安全防护这一对预测可靠性要求极高的关键应用领域而言,当前10%左右的误差水平仍显不足,模型精度亟待进一步提升以满足实际安全防护的严格需求。
针对多孔介质临界淬熄直径的高精度预测需求,系统对比了CNN与Transformer模型的预测性能。通过确定各模型最优超参数组合后,采用独立验证集进行鲁棒性评估,并可视化关键性能指标分布,如图8所示。定量分析表明:CNN模型在测试集上呈现δMAE=0.079、δMSE=0.012及R2=0.906;相较之下,Transformer模型展现出显著更优的统计特性——其δMAE=0.068和δMSE=0.008分别降低15.2%和33.3%,同时R2提升至0.928。该差异具有双重内涵:在误差维度,Transformer的δMAE/δMSE优势反映其对异常值更强的鲁棒性;在解释性维度,其更高的R2值证实模型捕获了更完备的物理机制非线性映射关系。进一步可推断Transformer的注意力机制有效聚焦于淬熄过程的临界特征参数(如孔隙率-流速耦合效应),而CNN的局部感知特性在全局工况泛化中存在结构性局限。综上,在当前以预测精度为核心指标的范式下,Transformer模型展现出明确的优势,这为复杂多孔介质燃爆安全防护提供了更可靠的算法基础。
为进一步探究模型预测精度的稳定性,系统对比了Transformer与CNN模型预测误差随真实值变化的分布特性,如图9所示。可视化分析表明,在临界淬熄直径的实测范围内,CNN模型的预测误差呈现出显著的系统性偏离。具体而言,其预测值与实验值间的绝对偏差在多个实测值区间均高于Transformer模型,尤其在峰值工况区域误差波动更剧烈。反观Transformer模型,其误差分布带明显低于CNN模型,且离散程度较低,表明该模型对不同工况的适应性更强。该现象印证了上述定量指标(δMAE/δMSE)的差异本质:CNN模型虽具备基础预测能力,但其误差分布的广泛性与不稳定性导致整体精度劣于Transformer。这一发现从误差结构维度揭示了Transformer在复杂非线性关系建模中的可靠性优势。
为了探究模型的推广维度,使用未归一化的数据评估模型的预测能力。在最优超参数组合条件下,对比了Transformer模型和CNN模型处理是否进行归一化处理的数据的性能差异。可以发现,源数据是否进行归一化处理对Transformer模型性能的影响很小,其δMAE和δMSE相差不大。但是,对于CNN模型,未进行归一化处理数据的R2为0.904。相比归一化处理后的数据降低了3.7%。说明,Transformer模型对于复杂数据的接受能力要强于CNN模型。换言之,对于未进行归一化处理的数据,Transformer模型也能获得令人满意的预测结果。
为验证模型在实际工业复杂数据中的泛化能力,系统评估了Transformer与CNN对原始尺度数据的适应性。在固定最优超参数条件下,设计2组对照模拟:分别输入归一化/未归一化数据2类模型。如图10所示,Transformer的归一化敏感性显著低于CNN。反观CNN模型,未归一化数据导致其决定系数由归一化状态的R2 = 0.906骤降至0.888。这种差异源于模型固有特性:Transformer的层归一化(LayerNorm)模块及自注意力机制具备隐式特征缩放能力,而CNN的卷积核权重受输入量纲直接影响。结果表明,Transformer在应对复杂工程数据(如含多物理量纲的临界淬熄直径数据集)时具有更强的鲁棒性,这对开发免预处理的实时安全防护系统具有重要应用价值。
基于燃气组分、管道几何参数、初始条件及多孔介质结构参数构成的九维特征空间,建立了密闭管道可燃气临界淬熄直径的预测模型。通过确定模型的最优超参数组合,系统对比CNN与Transformer模型的预测性能(δMAE、δMSE、 R2和误差分布)。研究表明Transformer模型预测性能显著优于CNN,且提高了对未归一化数据的鲁棒性,被确立为预测多孔介质临界淬熄直径的最优解。该模型突破了传统经验公式的工况局限,为管道燃爆安全防护体系提供核心算法支撑。得到的主要结论如下:
(1) 经最优超参数配置下的系统验证,CNN模型虽实现约10%的平均预测误差,但其显著存在的离群值及误差波动性表明,该模型在管道燃爆安全防护场景中仍存在可靠性缺陷。反观Transformer模型,在同等验证条件下将预测误差压缩至7%左右,且离群值比例显著降低。这种精度的阶跃式提升不仅证实了Transformer对多孔介质临界淬熄直径预测的强鲁棒性,更从本质上解决了安全防护领域的核心需求。Transformer模型误差分布的高度稳定性消除了传统CNN模型在极端工况下的误判风险,为燃气管道动态安全防控策略提供确定性保障。
(2) 基于多孔介质临界淬熄直径预测任务的系统验证,Transformer模型在最优超参数配置下展现出显著优于CNN的预测性能。其核心指标δMAE=0.068和δMSE=0.008较CNN(0.079和0.012)分别降低15.2%和33.3%,且相关系数R2提升至0.928(CNN的R2为0.906)。误差分布分析进一步证实,Transformer在全工况范围内保持窄带离散特性,这表明Transformer通过自注意力机制实现了对淬熄过程临界淬熄直径的精准预测,为安全阈值的可靠判定提供了算法基础。
(3) 在突破归一化预处理的泛化能力验证中,Transformer模型表现出卓越的尺度不变性,而CNN的R2则显著下降。该差异源于Transformer模型内置的层归一化机制与特征解耦能力,使其可直接处理多物理量纲混合的原始工程数据,对开发免预处理的实时安全防护系统具有重要应用价值。
评估了CNN与Transformer模型的预测性能,以确定预测多孔介质临界淬熄直径的最优模型。然而,使用的机器学习算法存在一定的局限性,还能够引入其他模型进行对比。因此,后续工作应利用更复杂的算法来构建多孔介质的关键参数预测模型。然后应将该模型与其他模型进行比较,以确定为最佳模型,并将预测结果进行分析。此外,气体爆炸的火焰淬熄数据集相对有限,需要扩展数据集以提高模型预测的精度。未来将对上述问题进行深入研究。

参考文献 引证文献
排序方式:
[1]
GROSSEL S S. Deflagration and detonation flame arresters [M]. America: Center for Chemical Process Safety of the American Institute of Chemical Engineers, 2002.
GROSSEL S S. Deflagration and detonation flame arresters [M]. America: Center for Chemical Process Safety of the American Institute of Chemical Engineers, 2002. DOI: 10.1002/9780470935651.
[2]
NIE Z H, GAO W, JIANG H P, et al. Predicting critical flame quenching thickness using machine learning approach with ResNet and ANN [J]. Journal of Loss Prevention in the Process Industries, 2024, 92: 105448. DOI: 10.1016/j.jlp.2024.105448.
[3]
LANGFORD B, PALMER K N, TONKIN P S. The performance of flame arresters against flames propagating in various fuel/air mixtures: Fire Research Note 486 [R]. Herts: Fire Research Station, 1961.
[4]
SONG X Z, ZUO X C, YANG Z K, et al. The explosion-suppression performance of mesh aluminum alloys and spherical nonmetallic materials on hydrogen-air mixtures [J]. International Journal of Hydrogen Energy, 2020, 45(56): 32686–32701. DOI: 10.1016/j.ijhydene.2020.08.197.
[5]
NIE B S, HE X Q, ZHANG R M, et al. The roles of foam ceramics in suppression of gas explosion overpressure and quenching of flame propagation [J]. Journal of Hazardous Materials, 2011, 192(2): 741–747. DOI: 10.1016/j.jhazmat.2011.05.083.
[6]
SUN S C, LIU G, LIU J X, et al. Effect of porosity and element thickness on flame quenching for in-line crimped-ribbon flame arresters [J]. Journal of Loss Prevention in the Process Industries, 2017, 50: 221–228. DOI: 10.1016/j.jlp.2017.09.017.
[7]
CAO X Y, WANG Z, WANG Y, et al. Effect of wire mesh structure parameter on the flame propagation characteristics of syngas explosion [J]. Fuel, 2023, 334: 126658. DOI: 10.1016/j.fuel.2022.126658.
[8]
LIU Z Y, GU Z Y, LIU P. Deep learning [M]//LIU Z Y, GU Z Y, LIU P. Transportation Big Data: Theory and Methods. Amsterdam: Elsevier, 2025: 383–430.
[9]
ZHAO F Y, GAO W, LU J X, et al. Real-time concentration detection of Al dust using GRU-based Kalman filtering approach [J]. Process Safety and Environmental Protection, 2024, 189: 154–163. DOI: 10.1016/j.psep.2024.06.052.
[10]
ZHANG K, ZHANG K, BAO R. Prediction of gas explosion pressures: a machine learning algorithm based on KPCA and an optimized LSSVM [J]. Journal of Loss Prevention in the Process Industries, 2023, 83: 105082. DOI: 10.1016/j.jlp.2023.105082.
[11]
CHEN J G, NIE Z H, ZHAO F Y, et al. Improving the stability of electrostatic induction dust concentration detection using kalman filtering algorithm aided by machine learning [J]. Process Safety and Environmental Protection, 2023, 174: 882–890. DOI: 10.1016/j.psep.2023.05.013.
[12]
GUIBERTI T F, BELHI M, DAMAZO J S, et al. Quenching distance of laminar methane-air flames at cryogenic temperatures and implications for flame arrester design [J]. Applications in Energy and Combustion Science, 2020, 1: 100001. DOI: 10.1016/j.jaecs.2020.100001.
[13]
FRIEDMAN R. The quenching of laminar oxyhydrogen flames by solid surfaces [J]. Symposium on Combustion and Flame, and Explosion Phenomena, 1948, 3(1): 110–120. DOI: 10.1016/S1062-2896(49)80014-6.
[14]
WEN X P, MA M, SUN W C, et al. The quenching characteristics of gas deflagration flame in narrow channel [J]. Advanced Materials Research, 2012, 455/456: 289–295. DOI: 10.4028/www.scientific.net/AMR.455-456.289.
[15]
WU Y, WEN X P, GUO Z D, et al. Experimental study on the propagation characteristics of hydrogen/methane/air premixed flames in a narrow channel [J]. International Journal of Hydrogen Energy, 2022, 47(9): 6377–6387. DOI: 10.1016/j.ijhydene.2021.12.019.
[16]
LONG F Y, DUAN Y L, YU S W, et al. Effect of porous materials on explosion characteristics of low ratio hydrogen/methane mixture in barrier tube [J]. Journal of Loss Prevention in the Process Industries, 2022, 80: 104875. DOI: 10.1016/j.jlp.2022.104875.
[17]
周凯元 李宗芬, 周自金. 波纹板阻火器对爆燃火焰淬熄作用的实验研究 [J]. 中国科学技术大学学报, 1997, 27(4): 77–82.
ZHOU K Y, LI Z F, ZHOU Z J. The quenching of deflagration by crimped-ribbon flame arresters [J]. Journal of China University of Science and Technology, 1997, 27(4): 77–82.
[18]
WANG L Q, MA H H, SHEN Z W. The quenching of propane deflagrations by crimped ribbon flame arrestors [J]. Journal of Loss Prevention in the Process Industries, 2016, 43: 567–574. DOI: 10.1016/j.jlp.2016.07.025.
[19]
HOLM J M. II. On the initiation of gaseous explosions by small flames [J]. The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science, 1932, 14(89): 18–56. DOI: 10.1080/14786443209462032.
[20]
PIOTROWSKI T C. Specification of flame arresting devices for manifolded low pressure storage tanks [J]. Plant/operations Progress, 1991, 10(2): 102–106. DOI: 10.1002/prsb.720100210.
[21]
CHEN D G, MA H H, SHEN Z W, et al. Investigation of the quenching of methane-air deflagration in narrow parallel channels [J]. Process Safety Progress, 2020, 39(3): e12126. DOI: 10.1002/prs.12126.
[22]
LV X S, YU J L, HOU Y J, et al. The quenching and attenuation of hydrogen-air detonation after passing across capillaries [J]. Fuel, 2022, 324: 124535. DOI: 10.1016/j.fuel.2022.124535.
[23]
SHI J H, ZHANG H, LI J J, et al. Real-time natural gas explosion modeling of offshore platforms by using deep learning probability approach [J]. Ocean Engineering, 2023, 276: 114244. DOI: 10.1016/j.oceaneng.2023.114244.
[24]
SUZUKI Y, NAKAYAMA J, SUZUKI T, et al. Consequence analysis of a small-scale hydrogen leakage from the overhead hydrogen piping based on machine learning and physical modeling [J]. Journal of Loss Prevention in the Process Industries, 2024, 90: 105328. DOI: 10.1016/j.jlp.2024.105328.
[25]
QIN G J, XIA A L, LU H F, et al. A hybrid machine learning model for predicting crater width formed by explosions of natural gas pipelines [J]. Journal of Loss Prevention in the Process Industries, 2023, 82: 104994. DOI: 10.1016/j.jlp.2023.104994.
[26]
XU Q M, CHEN G H, SU S, et al. Prediction of venting gas explosion overpressure based on a combination of explosive theory and machine learning [J]. Expert Systems with Applications, 2023, 234: 121044. DOI: 10.1016/j.eswa.2023.121044.
[27]
NIE Z H, GAO W, JIANG H P, et al. Quenching characteristics and mechanism of hydrogen-air mixtures by corrugated plate flame arrester under inert conditions [J]. Fuel, 2024, 362: 130822. DOI: 10.1016/j.fuel.2023.130822.
[28]
NIE Z H, GAO W, JIANG H P, et al. Explosion quenching characteristics of corrugated plate microchannels in ethanol-gasoline/air mixtures [J]. International Journal of Thermal Sciences, 2024, 204: 109220. DOI: 10.1016/j.ijthermalsci.2024.109220.
[29]
孙少辰. 波纹管道阻火器系统的阻爆特性研究 [D]. 大连: 大连理工大学, 2018.
SUN S C. Research of explosion resistant characteristic in in-line crimped-ribbon flame arrester [D]. Dalian: Dalian University of Technology, 2018.
[30]
万阳. 管道内针对氢-空气火焰的多孔板阻火特性研究 [D]. 合肥: 合肥工业大学, 2020. DOI: 10.27101/d.cnki.ghfgu.2020.001840.
WAN Y. Study on fire resistance characteristics of perforated plate for hydrogen-air flame in a channel [D]. Hefei: Hefei University of Technology, 2020. DOI: 10.27101/d.cnki.ghfgu.2020.001840.
[31]
张炼卓, 闫兴清, 吕先舒, 等. 不同初始压力下可爆预混气体通过波纹板阻火器的淬熄特性研究 [J]. 爆炸与冲击, 2023, 43(1): 012202. DOI: 10.11883/bzycj-2021-4058.
ZHANG L Z, YAN X Q, LV X S, et al. On quenching characteristics of combustible premixed gas through a crimped-ribbon flame arrester at different initial pressures [J]. Explosion and Shock Waves, 2023, 43(1): 012202. DOI: 10.11883/bzycj-2021-4058.
[32]
ZHONG X T, GALLAGHER B, EVES K, et al. A study of real-world micrograph data quality and machine learning model robustness [J]. NPJ Computational Materials, 2021, 7(1): 161. DOI: 10.1038/s41524-021-00616-3.
[33]
PEI X L, ZHAO Y H, CHEN L W, et al. Robustness of machine learning to color, size change, normalization, and image enhancement on micrograph datasets with large sample differences [J]. Materials & Design, 2023, 232: 112086. DOI: 10.1016/j.matdes.2023.112086.
[34]
VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need [C]//Proceedings of the 31st International Conference on Neural Information Processing Systems. Long Beach: Curran Associates Inc. , 2017: 6000–6010.
[35]
WAIBEL A, HANAZAWA T, HINTON G, et al. Phoneme recognition using time-delay neural networks [J]. IEEE Transactions on Acoustics, Speech, and Signal Processing, 1989, 37(3): 328–339. DOI: 10.1109/29.21701.
[36]
卢祖文. 基于X光图像与深度学习的耐张线夹缺陷识别 [D]. 南昌: 南昌大学, 2022. DOI: 10.27232/d.cnki.gnchu.2022.002699.
LU Z W. Recognition of tension clamp defects based on X-ray image and deep learning [D]. Nanchang: Nanchang University, 2022. DOI: 10.27232/d.cnki.gnchu.2022.002699.
[37]
BI Y B, WANG Y B, WANG S L, et al. A hybrid expert neural network for predicting hydrogen concentration under the ceiling in underground garage [J]. Building and Environment, 2025, 268: 112325. DOI: 10.1016/j.buildenv.2024.112325.
2026年第46卷第9期
PDF下载
129
62
引用本文
BibTeX
文章信息
doi: 10.11883/bzycj-2025-0218
  • 接收时间:2025-07-14
  • 首发时间:2026-10-10
  • 出版时间:2026-09-05
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2025-07-14
  • 修回日期:2026-03-28
基金
作者信息
    1大连理工大学化工学院,辽宁省 大连市 116024
    2北京航天试验技术研究所,北京 100074

通讯作者:

姜海鹏(1990- ),男,博士,副教授,
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/bzycj/CN/10.11883/bzycj-2025-0218
分享至
全文二维码

扫描看全文

引用本文
BibTeX
本文的引用情况
2种不同金属材料的力学参数
科
Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)
属
Genus
种数
Number of
species
占总种数比例
Percentage of total
species (%)
鹅膏菌科Amanitaceae 2 11 5.26 鹅膏菌属 Amanita 10 4.78
小菇科 Mycenaceae 2 12 5.74 丝盖伞属 Inocybe 5 2.39
多孔菌科 Polyporaceae 8 14 6.70 蜡蘑属 Laccaria 5 2.39
红菇科 Russulaceae 3 23 11.00 小皮伞属 Marasmius 6 2.87
小菇属 Mycena 11 5.26
光柄菇属 Pluteus 5 2.39
红菇属 Russula 17 8.13
栓菌属 Trametes 5 2.39
关闭全屏