Article(id=1194604063045165299, tenantId=1146029695717560320, journalId=1189918454225211397, issueId=1194604061094810310, articleNumber=null, orderNo=null, doi=10.20104/j.cnki.1674-6546.20240371, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=research-article, receivedDate=null, receivedDateStr=null, revisedDate=1739203200000, revisedDateStr=2025-02-11, acceptedDate=null, acceptedDateStr=null, onlineDate=1762745403299, onlineDateStr=2025-11-10, pubDate=1749916800000, pubDateStr=2025-06-15, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1762745403299, onlineIssueDateStr=2025-11-10, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1762745403299, creator=13701087609, updateTime=1762745403299, updator=13701087609, issue=Issue{id=1194604061094810310, tenantId=1146029695717560320, journalId=1189918454225211397, year='2025', volume='', issue='6', pageStart='1', pageEnd='48', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=null, createTime=1762745402821, creator=13701087609, updateTime=1762746663749, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1194609349818552989, tenantId=1146029695717560320, journalId=1189918454225211397, issueId=1194604061094810310, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1194609349818552990, tenantId=1146029695717560320, journalId=1189918454225211397, issueId=1194604061094810310, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=41, endPage=48, ext={EN=ArticleExt(id=1194604063296823541, articleId=1194604063045165299, tenantId=1146029695717560320, journalId=1189918454225211397, language=EN, title=Crash Simulation Optimization of Automotive Sill Beams Based on Machine Learning, columnId=1194604061958836937, journalTitle=Automotive Engineer, columnName=Special Topic on 2024 International Conference of Vehicle Safety and Intelligent Transportation, runingTitle=null, highlight=null, articleAbstract=

The resrearch aims of the paper is to improve the optimization efficiency of the automobile sill beam, and address the challenges of optimization such as the limited energy absorption of the sill extrusion aluminum under the condition of the side column collision, long iterative cycle of explicit solution in the simulation and the high requirements of the manufacturing process. A collision model is established based on an SUV model, and the Design Of Experiment (DOE) analysis is made with sill beam thickness as variables, resulting in 144 groups of valid design data are obtained. A Reduced-Order Model (ROM) is formed by deep learning methods(Rapidminer, romAI) and used as the alternative model of optimization and simulation. The results of CAE simulation verifies that accuracy of romAI reaches more than 95% in optimization and the solution speed is increased by more than 40 times under the limited data, which greatly shortens the R&D cycle.

, correspAuthors=null, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, 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, authorCompany=null, fund=null, authors=null, authorsList=Jiayuan Liu, Hongwei Peng, Guanrong Gu), CN=ArticleExt(id=1194604432844366419, articleId=1194604063045165299, tenantId=1146029695717560320, journalId=1189918454225211397, language=CN, title=基于机器学习的汽车门槛梁碰撞仿真优化, columnId=1194604062134997707, journalTitle=汽车工程师, columnName=2024年车辆安全与智能交通国际学术会议专题, runingTitle=null, highlight=null, articleAbstract=

为提高汽车门槛梁优化效率,针对汽车侧面柱碰撞工况门槛挤压铝吸能量有限、仿真中显式求解迭代周期长及制造工艺要求高等优化难点,基于某SUV车型建立碰撞模型,以门槛梁厚度为变量进行试验设计(DOE)分析,获得144组有效设计数据,通过深度学习方法(Rapidminer、romAI)形成降阶模型(ROM)用作优化仿真的替代模型,CAE仿真验证结果表明,在数据量有限的条件下,romAI在优化中的精度达到95%以上,且求解速度提高近40倍,极大缩短了研发周期。

, correspAuthors=null, authorNote=null, correspAuthorsNote=
刘家员(1991—),硕士研究生,工程师,主要研究方向为汽车被动安全,
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=FFbBV3JMTfXvqerBXGhrVA==, magXml=RDhZzm4Q0KjGN9AU8QK6Rg==, pdfUrl=null, pdf=QyZyT/X2ffSwl2CNgtQL6Q==, pdfFileSize=11366902, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=jgtoEpP7qLyUyiXdAJErlQ==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=eWXDoPVMvtJEqUjJWu55AA==, mapNumber=null, authorCompany=null, fund=null, authors=null, authorsList=刘家员, 彭竑维, 辜冠荣)}, authors=[Author(id=1194604433146356313, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=1391161812@qq.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1194604433213465179, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, authorId=1194604433146356313, language=EN, stringName=Jiayuan Liu, firstName=Jiayuan, middleName=null, lastName=Liu, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=Altair Engineering Software (Shanghai) Co., Ltd., Shanghai 200040, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1194604433272185436, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, authorId=1194604433146356313, language=CN, stringName=刘家员, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=澳汰尔工程软件(上海)有限公司, 上海 200040, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1194604433066664533, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, xref=null, ext=[AuthorCompanyExt(id=1194604433079247446, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, companyId=1194604433066664533, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Altair Engineering Software (Shanghai) Co., Ltd., Shanghai 200040), AuthorCompanyExt(id=1194604433087636055, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, companyId=1194604433066664533, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=澳汰尔工程软件(上海)有限公司, 上海 200040)])]), Author(id=1194604433330905694, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, 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=1194604433389625952, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, authorId=1194604433330905694, language=EN, stringName=Hongwei Peng, firstName=Hongwei, middleName=null, lastName=Peng, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=Altair Engineering Software (Shanghai) Co., Ltd., Shanghai 200040, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1194604433439957601, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, authorId=1194604433330905694, language=CN, stringName=彭竑维, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=澳汰尔工程软件(上海)有限公司, 上海 200040, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1194604433066664533, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, xref=null, ext=[AuthorCompanyExt(id=1194604433079247446, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, companyId=1194604433066664533, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Altair Engineering Software (Shanghai) Co., Ltd., Shanghai 200040), AuthorCompanyExt(id=1194604433087636055, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, companyId=1194604433066664533, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=澳汰尔工程软件(上海)有限公司, 上海 200040)])]), Author(id=1194604433507066467, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, 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=1194604433565786725, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, authorId=1194604433507066467, language=EN, stringName=Guanrong Gu, firstName=Guanrong, middleName=null, lastName=Gu, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=Altair Engineering Software (Shanghai) Co., Ltd., Shanghai 200040, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1194604433628701286, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, authorId=1194604433507066467, language=CN, stringName=辜冠荣, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=澳汰尔工程软件(上海)有限公司, 上海 200040, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1194604433066664533, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, xref=null, ext=[AuthorCompanyExt(id=1194604433079247446, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, companyId=1194604433066664533, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Altair Engineering Software (Shanghai) Co., Ltd., Shanghai 200040), AuthorCompanyExt(id=1194604433087636055, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, companyId=1194604433066664533, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=澳汰尔工程软件(上海)有限公司, 上海 200040)])])], keywords=[Keyword(id=1194604433708393063, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=EN, orderNo=1, keyword=Machine learning), Keyword(id=1194604433775501928, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=EN, orderNo=2, keyword=Reduced Order Model (ROM)), Keyword(id=1194604433855193705, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=EN, orderNo=3, keyword=Automobile sill beam), Keyword(id=1194604433922302570, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=EN, orderNo=4, keyword=Optimization simulations), Keyword(id=1194604433993605739, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=CN, orderNo=1, keyword=机器学习), Keyword(id=1194604434060714604, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=CN, orderNo=2, keyword=降阶模型), Keyword(id=1194604434127823469, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=CN, orderNo=3, keyword=汽车门槛梁), Keyword(id=1194604434199126638, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=CN, orderNo=4, keyword=优化仿真)], refs=[Reference(id=1194604438380847759, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2017, volume=327, issue=null, pageStart=327, pageEnd=351, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=OISHI A, YAGAWA G, journalName=Computer Methods in Applied Mechanics and Engineering, refType=null, unstructuredReference=OISHI A, YAGAWA G. Computational Mechanics Enhanced by Deep Learning[J]. Computer Methods in Applied Mechanics and Engineering, 2017, 327: 327-351., articleTitle=Computational Mechanics Enhanced by Deep Learning, refAbstract=null), Reference(id=1194604438473122448, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2022, volume=null, issue=null, pageStart=398, pageEnd=null, url=null, language=null, rfNumber=[2], rfOrder=1, authorNames=DESHPANDE S, LENGIEWICZ J, BORDAS S, journalName=Computer Methods in Applied Mechanics and Engineering, refType=null, unstructuredReference=DESHPANDE S, LENGIEWICZ J, BORDAS S. Probabilistic Deep Learning for Real-Time Large Deformation Simulations[J]. Computer Methods in Applied Mechanics and Engineering, 2022, 398., articleTitle=Probabilistic Deep Learning for Real-Time Large Deformation Simulations, refAbstract=null), Reference(id=1194604438540231313, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2018, volume=15, issue=138, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[3], rfOrder=2, authorNames=LIANG L, LIU M L, MARTIN C, journalName=Journal of the Royal Society Interface, refType=null, unstructuredReference=LIANG L, LIU M L, MARTIN C, et al. A Deep Learning Approach to Estimate Stress Distribution: A Fast and Accurate Surrogate of Finite-Element Analysis[J]. Journal of the Royal Society Interface, 2018, 15(138)., articleTitle=A Deep Learning Approach to Estimate Stress Distribution: A Fast and Accurate Surrogate of Finite-Element Analysis, refAbstract=null), Reference(id=1194604438619923090, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2019, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[4], rfOrder=3, authorNames=郭道一, journalName=基于深度学习和仿真的汽车碰撞损伤预测方法, refType=null, unstructuredReference=郭道一. 基于深度学习和仿真的汽车碰撞损伤预测方法[D]. 大连: 大连理工大学, 2019., articleTitle=null, refAbstract=null), Reference(id=1194604438708003475, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2019, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[4], rfOrder=4, authorNames=GUO D Y, journalName=Automobile Crash Damage Prediction Method Based on Deep Learning and Simulation, refType=null, unstructuredReference=GUO D Y. Automobile Crash Damage Prediction Method Based on Deep Learning and Simulation[D]. Dalian: Dalian University of Technology, 2019., articleTitle=null, refAbstract=null), Reference(id=1194604438775112340, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[5], rfOrder=5, authorNames=祝雪峰, 傅文谦, 郭道一, journalName=一种基于仿真数据和人工智能的汽车碰撞损伤等级实时预测方法: CN113094808A, refType=null, unstructuredReference=祝雪峰, 傅文谦, 郭道一. 一种基于仿真数据和人工智能的汽车碰撞损伤等级实时预测方法: CN113094808A[P]. 2021-07-09., articleTitle=null, refAbstract=null), Reference(id=1194604438842221205, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[5], rfOrder=6, authorNames=ZHU X F, FU W Q, GUO D Y, journalName=A Real-Time Prediction Method for Automobile Crash Damage Grade Based on Simulation Data and Artificial Intelligence: CN113094808A, refType=null, unstructuredReference=ZHU X F, FU W Q, GUO D Y. A Real-Time Prediction Method for Automobile Crash Damage Grade Based on Simulation Data and Artificial Intelligence: CN113094808A[P]. 2021-07-09., articleTitle=null, refAbstract=null), Reference(id=1194604438909330070, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[6], rfOrder=7, authorNames=高军, journalName=基于计算机视觉与深度学习的汽车主动安全系统关键技术研究, refType=null, unstructuredReference=高军. 基于计算机视觉与深度学习的汽车主动安全系统关键技术研究[D]. 武汉: 武汉理工大学, 2018., articleTitle=null, refAbstract=null), Reference(id=1194604438980633239, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[6], rfOrder=8, authorNames=GAO J, journalName=Research on Key Technologies of Automotive Active Safety System Based on Computer Vision and Deep Learning, refType=null, unstructuredReference=GAO J. Research on Key Technologies of Automotive Active Safety System Based on Computer Vision and Deep Learning[D]. Wuhan: Wuhan University of Technology, 2018., articleTitle=null, refAbstract=null), Reference(id=1194604439060325016, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=591, pageEnd=594, url=null, language=null, rfNumber=[7], rfOrder=9, authorNames=刘淼, 孙成智, 王潇嵩, journalName=2021中国汽车工程学会年会论文集, refType=null, unstructuredReference=刘淼, 孙成智, 王潇嵩. 基于深度学习的汽车结构件应力预测方法[C]// 2021中国汽车工程学会年会论文集. 上海: 机械工业出版社, 2021: 591-594., articleTitle=基于深度学习的汽车结构件应力预测方法, refAbstract=null), Reference(id=1194604439123239577, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=591, pageEnd=594, url=null, language=null, rfNumber=[7], rfOrder=10, authorNames=LIU M, SUN C Z, WANG X S, journalName=Proceedings of the 2021 Annual Conference of the Society of Automotive Engineers of China, refType=null, unstructuredReference=LIU M, SUN C Z, WANG X S. Stress Prediction Method for Automotive Structural Parts Based on Deep Learning[C]// Proceedings of the 2021 Annual Conference of the Society of Automotive Engineers of China. Shanghai: China Machine Press, 2021: 591-594., articleTitle=Stress Prediction Method for Automotive Structural Parts Based on Deep Learning, refAbstract=null), Reference(id=1194604439190348442, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2024, volume=46, issue=1, pageStart=170, pageEnd=178, url=null, language=null, rfNumber=[8], rfOrder=11, authorNames=高泽, 楚遵康, 石稼晟, journalName=汽车工程, refType=null, unstructuredReference=高泽, 楚遵康, 石稼晟, 等. 基于图网络的汽车零部件应力场快速预测方法研究[J]. 汽车工程, 2024, 46(1): 170-178., articleTitle=基于图网络的汽车零部件应力场快速预测方法研究, refAbstract=null), Reference(id=1194604439261651611, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2024, volume=46, issue=1, pageStart=170, pageEnd=178, url=null, language=null, rfNumber=[8], rfOrder=12, authorNames=GAO Z, CHU Z K, SHI J S, journalName=Automotive Engineering, refType=null, unstructuredReference=GAO Z, CHU Z K, SHI J S, et al. Research on Rapid Prediction Method of Stress Field of Auto Parts Based on Graph Network[J]. Automotive Engineering, 2024, 46(1): 170-178., articleTitle=Research on Rapid Prediction Method of Stress Field of Auto Parts Based on Graph Network, refAbstract=null), Reference(id=1194604439332954780, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[9], rfOrder=13, authorNames=王恩铭, journalName=基于深度学习与有限元方法的细观泡沫金属压缩力学性能研究, refType=null, unstructuredReference=王恩铭. 基于深度学习与有限元方法的细观泡沫金属压缩力学性能研究[D]. 长春: 吉林大学, 2024., articleTitle=null, refAbstract=null), Reference(id=1194604439416840861, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[9], rfOrder=14, authorNames=WANG E M, journalName=Study on Mechanical Properties of Mesoscopic Metal Foam Based on Deep Learning and Finite Element Method, refType=null, unstructuredReference=WANG E M. Study on Mechanical Properties of Mesoscopic Metal Foam Based on Deep Learning and Finite Element Method[D]. Changchun: Jilin University, 2024., articleTitle=null, refAbstract=null), Reference(id=1194604439488144030, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=null, pageStart=617, pageEnd=623, url=null, language=null, rfNumber=[10], rfOrder=15, authorNames=杨静, 张会, 冯旸赫, journalName=第八届中国指挥控制大会, refType=null, unstructuredReference=杨静, 张会, 冯旸赫, 等. 基于深度学习的作战仿真时序数据处理与分类问题研究[C]// 第八届中国指挥控制大会. 北京: 中国指挥与控制学会, 2020: 617-623., articleTitle=基于深度学习的作战仿真时序数据处理与分类问题研究, refAbstract=null), Reference(id=1194604439593001631, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=null, pageStart=617, pageEnd=623, url=null, language=null, rfNumber=[10], rfOrder=16, authorNames=YANG J, ZHANG H, FENG Y H, journalName=The 8th China Command and Control Conference, refType=null, unstructuredReference=YANG J, ZHANG H, FENG Y H, et al. Research on Time Series Data Processing and Classification of Combat Simulation Based on Deep Learning[C]// The 8th China Command and Control Conference. Beijing: Chinese Command and Control Society, 2020: 617-623., articleTitle=Research on Time Series Data Processing and Classification of Combat Simulation Based on Deep Learning, refAbstract=null), Reference(id=1194604439660110496, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[11], rfOrder=17, authorNames=杨叶倩, journalName=基于深度学习的碳纤维制备工艺双向优化与仿真算法研究, refType=null, unstructuredReference=杨叶倩. 基于深度学习的碳纤维制备工艺双向优化与仿真算法研究[D]. 济南: 山东大学, 2023., articleTitle=null, refAbstract=null), Reference(id=1194604439723025057, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[11], rfOrder=18, authorNames=YANG Y Q, journalName=Research on Two-Way Optimization and Simulation Algorithm of Carbon Fiber Preparation Process Based on Deep Learning, refType=null, unstructuredReference=YANG Y Q. Research on Two-Way Optimization and Simulation Algorithm of Carbon Fiber Preparation Process Based on Deep Learning[D]. Jinan: Shandong University, 2023., articleTitle=null, refAbstract=null), Reference(id=1194604439785939618, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=4, pageStart=8, pageEnd=11, url=null, language=null, rfNumber=[12], rfOrder=19, authorNames=黄泽辉, 唐洪斌, 王雪松, journalName=汽车工程师, refType=null, unstructuredReference=黄泽辉, 唐洪斌, 王雪松, 等. 基于深度学习的车辆碰撞乘员伤害预测方法研究[J]. 汽车工程师, 2024(4): 8-11., articleTitle=基于深度学习的车辆碰撞乘员伤害预测方法研究, refAbstract=null), Reference(id=1194604439853048483, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=4, pageStart=8, pageEnd=11, url=null, language=null, rfNumber=[12], rfOrder=20, authorNames=HUANG Z H, TANG H B, WANG X S, journalName=Automotive Engineer, refType=null, unstructuredReference=HUANG Z H, TANG H B, WANG X S, et al. Research on Vehicle Collision Occupant Injury Prediction Method Based on Deep Learning[J]. Automotive Engineer, 2024(4): 8-11., articleTitle=Research on Vehicle Collision Occupant Injury Prediction Method Based on Deep Learning, refAbstract=null), Reference(id=1194604439928545956, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2024, volume=33, issue=4, pageStart=21, pageEnd=24, url=null, language=null, rfNumber=[13], rfOrder=21, authorNames=刘家员, 冷川, journalName=计算机辅助工程, refType=null, unstructuredReference=刘家员, 冷川. 基于汽车架构子模型的快速仿真方法[J]. 计算机辅助工程, 2024, 33(4): 21-24., articleTitle=基于汽车架构子模型的快速仿真方法, refAbstract=null), Reference(id=1194604440012432037, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2024, volume=33, issue=4, pageStart=21, pageEnd=24, url=null, language=null, rfNumber=[13], rfOrder=22, authorNames=LIU J Y, LENG C, journalName=Computer Aided Engineering, refType=null, unstructuredReference=LIU J Y, LENG C. Fast Simulation Method Based on Submodel of Vehicle Architecture[J]. Computer Aided Engineering, 2024, 33(4): 21-24., articleTitle=Fast Simulation Method Based on Submodel of Vehicle Architecture, refAbstract=null), Reference(id=1194604440108901030, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2012, volume=null, issue=49, pageStart=24, pageEnd=26, url=null, language=null, rfNumber=[14], rfOrder=23, authorNames=唐波, 赵晓红, journalName=汽车与配件, refType=null, unstructuredReference=唐波, 赵晓红. 汽车侧面柱碰撞的研究以及法规动态[J]. 汽车与配件, 2012(49): 24-26., articleTitle=汽车侧面柱碰撞的研究以及法规动态, refAbstract=null), Reference(id=1194604440171815591, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2012, volume=null, issue=49, pageStart=24, pageEnd=26, url=null, language=null, rfNumber=[14], rfOrder=24, authorNames=TANG B, ZHAO X H, journalName=Auto & Accessories, refType=null, unstructuredReference=TANG B, ZHAO X H. Research on Side Pillar Collision of Automobile and Regulatory Dynamics[J]. Auto & Accessories, 2012(49): 24-26., articleTitle=Research on Side Pillar Collision of Automobile and Regulatory Dynamics, refAbstract=null), Reference(id=1194604440234730152, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[15], rfOrder=25, authorNames=范亚彬, journalName=汽车侧面柱碰撞试验装置设计及仿真优化, refType=null, unstructuredReference=范亚彬. 汽车侧面柱碰撞试验装置设计及仿真优化[D]. 天津: 河北工业大学, 2017., articleTitle=null, refAbstract=null), Reference(id=1194604441291694761, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[15], rfOrder=26, authorNames=FAN Y B, journalName=Design and Simulation Optimization of Automobile Side Column Crash Test Device, refType=null, unstructuredReference=FAN Y B. Design and Simulation Optimization of Automobile Side Column Crash Test Device[D]. Tianjin: Hebei University of Technology, 2017., articleTitle=null, refAbstract=null), Reference(id=1194604441367192234, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[16], rfOrder=27, authorNames=李凯, journalName=纯电动汽车动力锂电池箱底部碰撞防护结构设计与仿真分析, refType=null, unstructuredReference=李凯. 纯电动汽车动力锂电池箱底部碰撞防护结构设计与仿真分析[D]. 长沙: 中南大学, 2023., articleTitle=null, refAbstract=null), Reference(id=1194604441430106795, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[16], rfOrder=28, authorNames=LI K, journalName=Design and Simulation Analysis of Collision Protection Structure at the Bottom of Power Lithium Battery Box of Pure Electric Vehicle, refType=null, unstructuredReference=LI K. Design and Simulation Analysis of Collision Protection Structure at the Bottom of Power Lithium Battery Box of Pure Electric Vehicle[D]. Changsha: Central South University, 2023., articleTitle=null, refAbstract=null), Reference(id=1194604441501409964, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2022, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[17], rfOrder=29, authorNames=朱啟兵, 吴业波, 曹雄, journalName=防护性强的电动车锂电池用保护板支架: CN216488291U, refType=null, unstructuredReference=朱啟兵, 吴业波, 曹雄, 等. 防护性强的电动车锂电池用保护板支架: CN216488291U[P]. 2022-05-10., articleTitle=null, refAbstract=null), Reference(id=1194604441572713133, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2022, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[17], rfOrder=30, authorNames=ZHU Q B, WU Y B, CAO X, journalName=Protection Plate Bracket for Lithium Battery of Electric Vehicle with Strong Protection: CN216488291U, refType=null, unstructuredReference=ZHU Q B, WU Y B, CAO X, et al. Protection Plate Bracket for Lithium Battery of Electric Vehicle with Strong Protection: CN216488291U[P]. 2022-05-10., articleTitle=null, refAbstract=null), Reference(id=1194604441635627694, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2005, volume=21, issue=3, pageStart=350, pageEnd=367, url=null, language=null, rfNumber=[18], rfOrder=31, authorNames=KURKOVA V, SANGUINETI M, journalName=Journal of Complexity, refType=null, unstructuredReference=KURKOVA V, SANGUINETI M. Learning with Generalization Capability by Kernel Methods of Bounded Complexity[J]. Journal of Complexity, 2005, 21(3): 350-367., articleTitle=Learning with Generalization Capability by Kernel Methods of Bounded Complexity, refAbstract=null), Reference(id=1194604441706930863, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2007, volume=31, issue=7, pageStart=1271, pageEnd=1281, url=null, language=null, rfNumber=[19], rfOrder=32, authorNames=ZHAO Z Q, HUANG D S, journalName=Applied Mathematical Modelling, refType=null, unstructuredReference=ZHAO Z Q, HUANG D S. A Mended Hybrid Learning Algorithm for Radial Basis Function Neural Networks to Improve Generalization Capability[J]. Applied Mathematical Modelling, 2007, 31(7): 1271-1281., articleTitle=A Mended Hybrid Learning Algorithm for Radial Basis Function Neural Networks to Improve Generalization Capability, refAbstract=null), Reference(id=1194604441774039728, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=1999, volume=107, issue=2, pageStart=347, pageEnd=356, url=null, language=null, rfNumber=[20], rfOrder=33, authorNames=LEUNG C T, CHOW T W S, journalName=Artificial Intelligence, refType=null, unstructuredReference=LEUNG C T, CHOW T W S. Adaptive Regularization Parameter Selection Method for Enhancing Generalization Capability of Neural Networks[J]. Artificial Intelligence, 1999, 107(2): 347-356., articleTitle=Adaptive Regularization Parameter Selection Method for Enhancing Generalization Capability of Neural Networks, refAbstract=null), Reference(id=1194604441845342897, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2000, volume=38, issue=10, pageStart=1853, pageEnd=1862, url=null, language=null, rfNumber=[21], rfOrder=34, authorNames=HALL K C, THOMAS J P, DOWELL E H, journalName=AIAA Journal, refType=null, unstructuredReference=HALL K C, THOMAS J P, DOWELL E H. Proper Orthogonal Decomposition Technique for Transonic Unsteady Aerodynamic Flows[J]. AIAA Journal, 2000, 38(10): 1853-1862., articleTitle=Proper Orthogonal Decomposition Technique for Transonic Unsteady Aerodynamic Flows, refAbstract=null), Reference(id=1194604441908257458, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2012, volume=39, issue=8, pageStart=1417, pageEnd=1429, url=null, language=null, rfNumber=[22], rfOrder=35, authorNames=RAVEH D E, journalName=AIAA Journal, refType=null, unstructuredReference=RAVEH D E. Reduced-Order Models for Nonlinear Unsteady Aerodynamics[J]. AIAA Journal, 2012, 39(8): 1417-1429., articleTitle=Reduced-Order Models for Nonlinear Unsteady Aerodynamics, refAbstract=null), Reference(id=1194604441983754931, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2013, volume=35, issue=8, pageStart=1798, pageEnd=1828, url=null, language=null, rfNumber=[23], rfOrder=36, authorNames=BENGIO Y, COURVILLE A, VINCENT P, journalName=IEEE Transactions on Pattern Analysis & Machine Intelligence, refType=null, unstructuredReference=BENGIO Y, COURVILLE A, VINCENT P. Representation Learning: A Review and New Perspectives[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2013, 35(8): 1798-1828., articleTitle=Representation Learning: A Review and New Perspectives, refAbstract=null), Reference(id=1194604442063446708, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2015, volume=6, issue=3, pageStart=261, pageEnd=270, url=null, language=null, rfNumber=[24], rfOrder=37, authorNames=寇家庆, 张伟伟, journalName=航空工程进展, refType=null, unstructuredReference=寇家庆, 张伟伟. 基函数宽度对递归RBF神经网络气动力模型精度的影响研究[J]. 航空工程进展, 2015, 6(3): 261-270., articleTitle=基函数宽度对递归RBF神经网络气动力模型精度的影响研究, refAbstract=null), Reference(id=1194604442134749877, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2015, volume=6, issue=3, pageStart=261, pageEnd=270, url=null, language=null, rfNumber=[24], rfOrder=38, authorNames=KOU J Q, ZHANG W W, journalName=Advancesin Aeronautical Scienceand Engineering, refType=null, unstructuredReference=KOU J Q, ZHANG W W. Research on the Effects of Basis Function Widths of Aerodynamic Modeling Based on Recursive RBF Neural Network[J]. Advancesin Aeronautical Scienceand Engineering, 2015, 6(3): 261-270., articleTitle=Research on the Effects of Basis Function Widths of Aerodynamic Modeling Based on Recursive RBF Neural Network, refAbstract=null), Reference(id=1194604442201858742, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=2012, volume=48, issue=10, pageStart=2418, pageEnd=2429, url=null, language=null, rfNumber=[25], rfOrder=39, authorNames=GLAZ B, LIU L, FRIEDMANN P P, journalName=AIAA Journal, refType=null, unstructuredReference=GLAZ B, LIU L, FRIEDMANN P P. Reduced-Order Nonlinear Unsteady Aerodynamic Modeling Using a Surrogate-Based Recurrence Framework[J]. AIAA Journal, 2012, 48(10): 2418-2429., articleTitle=Reduced-Order Nonlinear Unsteady Aerodynamic Modeling Using a Surrogate-Based Recurrence Framework, refAbstract=null), Reference(id=1194604442273161911, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, doi=null, pmid=null, pmcid=null, year=1997, volume=11, issue=4, pageStart=341, pageEnd=359, url=null, language=null, rfNumber=[26], rfOrder=40, authorNames=STORN R, PRICE K, journalName=Journal of Global Optimization, refType=null, unstructuredReference=STORN R, PRICE K. Differential Evolution-A Simple and Efficient Heuristic for Global Optimization over Continuous Spaces[J]. Journal of Global Optimization, 1997, 11(4): 341-359., articleTitle=Differential Evolution-A Simple and Efficient Heuristic for Global Optimization over Continuous Spaces, refAbstract=null)], funds=null, companyList=[AuthorCompany(id=1194604433066664533, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, xref=null, ext=[AuthorCompanyExt(id=1194604433079247446, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, companyId=1194604433066664533, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Altair Engineering Software (Shanghai) Co., Ltd., Shanghai 200040), AuthorCompanyExt(id=1194604433087636055, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, companyId=1194604433066664533, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=澳汰尔工程软件(上海)有限公司, 上海 200040)])], figs=[ArticleFig(id=1194604434383676015, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=EN, label=null, caption=null, figureFileSmall=i8zBuyOLQxadFIsIlB8OhA==, figureFileBig=jgtoEpP7qLyUyiXdAJErlQ==, tableContent=null), ArticleFig(id=1194604434488533616, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=CN, label=图1, caption=CAE仿真模型, figureFileSmall=i8zBuyOLQxadFIsIlB8OhA==, figureFileBig=jgtoEpP7qLyUyiXdAJErlQ==, tableContent=null), ArticleFig(id=1194604434656305777, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=EN, label=null, caption=null, figureFileSmall=89Km6DOw+oNx5FKO7Hln0Q==, figureFileBig=N1uvYQHRrSYbLpLZbrFnSw==, tableContent=null), ArticleFig(id=1194604434723414642, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=CN, label=图2, caption=侵入量测量弹簧, figureFileSmall=89Km6DOw+oNx5FKO7Hln0Q==, figureFileBig=N1uvYQHRrSYbLpLZbrFnSw==, tableContent=null), ArticleFig(id=1194604434803106419, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=EN, label=null, caption=null, figureFileSmall=UkBL5zA9+TMgbjYscx/sSw==, figureFileBig=HvxZgN5fGodvoRy1O1EsBg==, tableContent=null), ArticleFig(id=1194604434891186804, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=CN, label=图3, caption=柱碰撞工况门槛侵入量和门槛挤压力响应曲线, figureFileSmall=UkBL5zA9+TMgbjYscx/sSw==, figureFileBig=HvxZgN5fGodvoRy1O1EsBg==, tableContent=null), ArticleFig(id=1194604434958295669, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=EN, label=null, caption=null, figureFileSmall=EKG7oAHcBCJ6Tg4RUCeTrQ==, figureFileBig=Wsk8T1v9UX5nQfKnPOejMw==, tableContent=null), ArticleFig(id=1194604435008627318, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=CN, label=图4, caption=门槛梁优化流程, figureFileSmall=EKG7oAHcBCJ6Tg4RUCeTrQ==, figureFileBig=Wsk8T1v9UX5nQfKnPOejMw==, tableContent=null), ArticleFig(id=1194604435063153271, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=EN, label=null, caption=null, figureFileSmall=Ks1yCX2Y2RnvUK+Mv3HCiQ==, figureFileBig=MT9t17cpgJxrD5CCcBSy8g==, tableContent=null), ArticleFig(id=1194604435126067832, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=CN, label=图5, caption=门槛梁厚度DOE样本数据, figureFileSmall=Ks1yCX2Y2RnvUK+Mv3HCiQ==, figureFileBig=MT9t17cpgJxrD5CCcBSy8g==, tableContent=null), ArticleFig(id=1194604435188982393, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=EN, label=null, caption=null, figureFileSmall=zWW4Qg8rRSFJyY6a2cIuyw==, figureFileBig=UoxZyAty5VrzuKTxNHzNuQ==, tableContent=null), ArticleFig(id=1194604435256091258, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=CN, label=图6, caption=Knowledge Studio深度学习过程, figureFileSmall=zWW4Qg8rRSFJyY6a2cIuyw==, figureFileBig=UoxZyAty5VrzuKTxNHzNuQ==, tableContent=null), ArticleFig(id=1194604435331588731, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=EN, label=null, caption=null, figureFileSmall=yPXq1TZYKBY4jhMugKEOLA==, figureFileBig=RIW3cytpzxWzDcgwtADZNA==, tableContent=null), ArticleFig(id=1194604435402891900, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=CN, label=图7, caption=romAI模型映射关系, figureFileSmall=yPXq1TZYKBY4jhMugKEOLA==, figureFileBig=RIW3cytpzxWzDcgwtADZNA==, tableContent=null), ArticleFig(id=1194604435474195069, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=EN, label=null, caption=null, figureFileSmall=Q2f0bVe8Upe0jMerC11Ksg==, figureFileBig=q8Tz/oiBThBLjnyGh6dFxg==, tableContent=null), ArticleFig(id=1194604435633578622, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=CN, label=图8, caption=romAI降阶模型前处理, figureFileSmall=Q2f0bVe8Upe0jMerC11Ksg==, figureFileBig=q8Tz/oiBThBLjnyGh6dFxg==, tableContent=null), ArticleFig(id=1194604435709076095, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=EN, label=null, caption=null, figureFileSmall=by08mirS9/dnBBdQo3ajqw==, figureFileBig=5Q0RoyvB0liY7vwgJHviNg==, tableContent=null), ArticleFig(id=1194604436824760960, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=CN, label=图9, caption=深度学习精度匹配, figureFileSmall=by08mirS9/dnBBdQo3ajqw==, figureFileBig=5Q0RoyvB0liY7vwgJHviNg==, tableContent=null), ArticleFig(id=1194604436887675521, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=EN, label=null, caption=null, figureFileSmall=R0TBSt1oNETADqyZp53cjQ==, figureFileBig=Hc8XyNmXxMQxTF5S0yTlOA==, tableContent=null), ArticleFig(id=1194604436954784386, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=CN, label=图10, caption=romAI降阶模型, figureFileSmall=R0TBSt1oNETADqyZp53cjQ==, figureFileBig=Hc8XyNmXxMQxTF5S0yTlOA==, tableContent=null), ArticleFig(id=1194604437021893251, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=EN, label=null, caption=null, figureFileSmall=btZmuU2xrF1jOFA6dcfXag==, figureFileBig=O8dAgaluhYNaTG+f/afwYQ==, tableContent=null), ArticleFig(id=1194604437126750852, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=CN, label=图11, caption=柱碰撞门槛梁侵入量和挤压力对比, figureFileSmall=btZmuU2xrF1jOFA6dcfXag==, figureFileBig=O8dAgaluhYNaTG+f/afwYQ==, tableContent=null), ArticleFig(id=1194604437281940101, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=EN, label=null, caption=null, figureFileSmall=null, figureFileBig=null, tableContent=
模型宽度 模型深度 侵入量精度 挤压力精度
训练 测试 训练 测试
50 3 0.997 0.463 0.976 0.172
10 1 0.922 0.832 0.967 0.687
3 2 0.920 0.843 0.862 0.809
5 2 0.974 0.674 0.971 0.406
7 2 0.987 0.648 0.984 0.581
), ArticleFig(id=1194604437349048966, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=CN, label=表1, caption=

机器学习精度对比

, figureFileSmall=null, figureFileBig=null, tableContent=
模型宽度 模型深度 侵入量精度 挤压力精度
训练 测试 训练 测试
50 3 0.997 0.463 0.976 0.172
10 1 0.922 0.832 0.967 0.687
3 2 0.920 0.843 0.862 0.809
5 2 0.974 0.674 0.971 0.406
7 2 0.987 0.648 0.984 0.581
), ArticleFig(id=1194604437432935047, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=EN, label=null, caption=null, figureFileSmall=null, figureFileBig=null, tableContent=
优化方式 模型宽度 模型深度 质量/kg 侵入量 挤压力
优化结果/mm 验证结果/mm 误差/% 优化结果/kN 验证结果/kN 误差/%
GRSM 50 3 24.8 151.2 202.8 -25.4 235.8 229.6 2.7
GRSM 10 1 25.9 154.5 184.2 -16.1 261.0 233.3 11.9
GRSM 3 2 26.6 155.2 166.2 -6.6 255.3 246.5 3.5
SQP 50 3 23.9 155.3 220.4 -29.5 208.1 220.0 -5.4
SQP 10 1 25.4 155.0 193.1 -19.7 254.1 227.3 11.8
SQP 3 2 26.2 155.0 166.2 -6.8 246.7 246.7 0.0
), ArticleFig(id=1194604437525209736, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=CN, label=表2, caption=

机器学习后参数优化对比

, figureFileSmall=null, figureFileBig=null, tableContent=
优化方式 模型宽度 模型深度 质量/kg 侵入量 挤压力
优化结果/mm 验证结果/mm 误差/% 优化结果/kN 验证结果/kN 误差/%
GRSM 50 3 24.8 151.2 202.8 -25.4 235.8 229.6 2.7
GRSM 10 1 25.9 154.5 184.2 -16.1 261.0 233.3 11.9
GRSM 3 2 26.6 155.2 166.2 -6.6 255.3 246.5 3.5
SQP 50 3 23.9 155.3 220.4 -29.5 208.1 220.0 -5.4
SQP 10 1 25.4 155.0 193.1 -19.7 254.1 227.3 11.8
SQP 3 2 26.2 155.0 166.2 -6.8 246.7 246.7 0.0
), ArticleFig(id=1194604437625873033, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=EN, label=null, caption=null, figureFileSmall=null, figureFileBig=null, tableContent=
门槛梁加强筋厚度/mm 质量/kg 侵入量/mm 挤压力/kN
加强筋1 加强筋2 加强筋3 加强筋4 加强筋5
3.4 3.2 2.1 2.7 2.6 25.59 -150.77 256.32
3.4 3.1 2.0 2.6 2.2 25.06 -151.91 246.88
3.4 3.1 2.0 2.6 2.1 24.89 -153.39 243.42
3.4 3.1 2.0 2.5 2.1 24.82 -155.35 244.61
3.4 3.1 2.0 2.5 2.1 24.82 -155.35 244.61
3.4 3.1 2.0 2.5 2.1 24.82 -155.35 244.61
3.4 3.1 2.0 2.5 2.1 24.82 -155.35 244.61
3.4 3.1 2.0 2.5 2.1 24.82 -155.35 244.61
3.4 3.1 2.0 2.5 2.1 24.82 -155.35 244.61
3.4 3.1 2.0 2.5 2.1 24.82 -155.35 244.61
3.4 3.1 2.0 2.5 2.1 24.82 -155.35 244.61
3.3 2.9 2.0 3.1 2.0 24.76 -151.21 235.75
), ArticleFig(id=1194604437701370506, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=CN, label=表3, caption=

HyperStudy最优解

, figureFileSmall=null, figureFileBig=null, tableContent=
门槛梁加强筋厚度/mm 质量/kg 侵入量/mm 挤压力/kN
加强筋1 加强筋2 加强筋3 加强筋4 加强筋5
3.4 3.2 2.1 2.7 2.6 25.59 -150.77 256.32
3.4 3.1 2.0 2.6 2.2 25.06 -151.91 246.88
3.4 3.1 2.0 2.6 2.1 24.89 -153.39 243.42
3.4 3.1 2.0 2.5 2.1 24.82 -155.35 244.61
3.4 3.1 2.0 2.5 2.1 24.82 -155.35 244.61
3.4 3.1 2.0 2.5 2.1 24.82 -155.35 244.61
3.4 3.1 2.0 2.5 2.1 24.82 -155.35 244.61
3.4 3.1 2.0 2.5 2.1 24.82 -155.35 244.61
3.4 3.1 2.0 2.5 2.1 24.82 -155.35 244.61
3.4 3.1 2.0 2.5 2.1 24.82 -155.35 244.61
3.4 3.1 2.0 2.5 2.1 24.82 -155.35 244.61
3.3 2.9 2.0 3.1 2.0 24.76 -151.21 235.75
), ArticleFig(id=1194604437802033803, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=EN, label=null, caption=null, figureFileSmall=null, figureFileBig=null, tableContent=
对比参数 目标 基础结果 优化结果(GRSM) 验证结果 误差/%
质量/kg 最小 27.25 26.9 26.9 0
最大侵入量/mm <155 192.3 150.7 152.3 -1.0
最大挤压力/kN <250 236.6 255.5 250.6 2.0
), ArticleFig(id=1194604438049497740, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=CN, label=表4, caption=

优化后结果对比

, figureFileSmall=null, figureFileBig=null, tableContent=
对比参数 目标 基础结果 优化结果(GRSM) 验证结果 误差/%
质量/kg 最小 27.25 26.9 26.9 0
最大侵入量/mm <155 192.3 150.7 152.3 -1.0
最大挤压力/kN <250 236.6 255.5 250.6 2.0
), ArticleFig(id=1194604438196298381, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=EN, label=null, caption=null, figureFileSmall=null, figureFileBig=null, tableContent=
优化
方法
软件
应用
模型
宽度
模型
深度
质量/kg 侵入量 挤压力
优化结果
/mm
验证结果
/mm
误差/% 优化结果
/kN
验证结果
/kN
误差/%
GRSM KS 50 3 24.8 151.2 202.8 -25.4 235.8 229.6 2.7
romAI 50 3 26.9 155.2 152.3 1.9 260.9 250.6 4.1
KS 10 1 25.9 154.5 184.2 -16.1 261.0 233.3 11.9
KS 3 2 26.6 155.2 166.2 -6.6 255.3 246.5 3.5
romAI 120 5 26.9 150.7 152.3 -1.0 255.5 250.6 2.0
SQP KS 50 3 23.9 155.3 220.4 -29.5 208.1 220.0 -5.4
KS 10 1 25.4 155.0 193.1 -19.7 254.1 227.3 11.8
KS 3 2 26.2 155.0 166.2 -6.8 246.7 246.7 0.0
), ArticleFig(id=1194604438271795854, tenantId=1146029695717560320, journalId=1189918454225211397, articleId=1194604063045165299, language=CN, label=表5, caption=

深度学习的优化结果对比

, figureFileSmall=null, figureFileBig=null, tableContent=
优化
方法
软件
应用
模型
宽度
模型
深度
质量/kg 侵入量 挤压力
优化结果
/mm
验证结果
/mm
误差/% 优化结果
/kN
验证结果
/kN
误差/%
GRSM KS 50 3 24.8 151.2 202.8 -25.4 235.8 229.6 2.7
romAI 50 3 26.9 155.2 152.3 1.9 260.9 250.6 4.1
KS 10 1 25.9 154.5 184.2 -16.1 261.0 233.3 11.9
KS 3 2 26.6 155.2 166.2 -6.6 255.3 246.5 3.5
romAI 120 5 26.9 150.7 152.3 -1.0 255.5 250.6 2.0
SQP KS 50 3 23.9 155.3 220.4 -29.5 208.1 220.0 -5.4
KS 10 1 25.4 155.0 193.1 -19.7 254.1 227.3 11.8
KS 3 2 26.2 155.0 166.2 -6.8 246.7 246.7 0.0
)], attaches=null, journal=Journal(id=1189918244568731652, delFlag=0, nameCn=汽车工程师, nameEn=Automotive Engineer, nameHistory1=null, nameHistory2=null, issn=1674-6546, eissn=null, cn=22-1432/U, 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=+bJsKkKt/pjz9u6EwhnksQ==, journalPrice=null, startedYear=null, abbrevIsoEn=null, journalRemark=null, publicationField=null, createdTime=1761628217121, updatedTime=1761735708780, createdBy=18614031015, updatedBy=13701087609, firstLetterCn=A, firstLetterEn=A, subjectCode=Engineering, subjectName=Engineering, subjectCodeEn=Engineering, subjectNameEn=null, picCn=+bJsKkKt/pjz9u6EwhnksQ==, picEn=O3Sn3tnYYrh/jm6emnnMWA==, jcr=null, cjcr=null, exts=[JournalExt(id=1190369097415233706, language=CN, name=汽车工程师, 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=1761735708812, updatedTime=1761735708812, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://tjqc.cbpt.cnki.net/index.aspx?t=1, submissionEditorUrl=https://tjqc.cbpt.cnki.net/index.aspx?t=3, submissionReviewUrl=https://tjqc.cbpt.cnki.net/index.aspx?t=2, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1190369097553645739, language=EN, name=Automotive Engineer, 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=1761735708845, updatedTime=1761735708845, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://tjqc.cbpt.cnki.net/index.aspx?t=1, submissionEditorUrl=https://tjqc.cbpt.cnki.net/index.aspx?t=3, submissionReviewUrl=https://tjqc.cbpt.cnki.net/index.aspx?t=2, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1189918454225211397, websiteList=[Website(id=1189918982430847716, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1189918454225211397, 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/qcgcs/CN, language=CN, createTime=1761628393037, createBy=18614031015, updateTime=1761628422913, updateBy=18614031015, name=汽车工程师-中文, tplId=1146099689490845704, title=汽车工程师, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1189919800185917791, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982430847716, code=articleTextType, value=kx, createTime=1761628588005, updateTime=1761628588005, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919800164946268, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982430847716, code=banner, value=null, createTime=1761628588000, updateTime=1761628588000, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919800211083618, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982430847716, code=grayFlag, value=0, createTime=1761628588011, updateTime=1761628588011, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919800156557659, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982430847716, code=logo, value=https://castjournals.cast.org.cn/joweb/qcgcs/CN/file/pic?fileId=yiZ96RYoYcnGnRMuWdmkWA==, createTime=1761628587998, updateTime=1761628587998, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919800223666532, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982430847716, code=minRunFlag, value=0, createTime=1761628588014, updateTime=1761628588014, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919800181723486, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982430847716, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/qcgcs/CN/file/pic, createTime=1761628588004, updateTime=1761628588004, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919800215277923, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982430847716, code=silenceFlag, value=0, createTime=1761628588012, updateTime=1761628588012, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919800173334877, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982430847716, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1761628588002, updateTime=1761628588002, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919800194306400, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982430847716, code=themeColor, value=null, createTime=1761628588007, updateTime=1761628588007, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919800202695009, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982430847716, code=themeStyle, value=null, createTime=1761628588009, updateTime=1761628588009, creator=18614031015, updator=18614031015)]), Website(id=1189918982527316711, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1189918454225211397, 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/qcgcs/EN, language=EN, createTime=1761628393061, createBy=18614031015, updateTime=1761628543075, updateBy=18614031015, name=汽车工程师-英文, tplId=1146101810881728533, title=Automotive Engineer, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1189919837561352952, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982527316711, code=articleTextType, value=kx, createTime=1761628596916, updateTime=1761628596916, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919837540381429, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982527316711, code=banner, value=null, createTime=1761628596911, updateTime=1761628596911, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919837582324475, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982527316711, code=grayFlag, value=0, createTime=1761628596921, updateTime=1761628596921, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919837527798516, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982527316711, code=logo, value=https://castjournals.cast.org.cn/joweb/qcgcs/EN/file/pic?fileId=yiZ96RYoYcnGnRMuWdmkWA==, createTime=1761628596908, updateTime=1761628596908, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919837594907389, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982527316711, code=minRunFlag, value=0, createTime=1761628596924, updateTime=1761628596924, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919837557158647, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982527316711, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/qcgcs/EN/file/pic, createTime=1761628596915, updateTime=1761628596915, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919837586518780, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982527316711, code=silenceFlag, value=0, createTime=1761628596922, updateTime=1761628596922, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919837548770038, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982527316711, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1761628596913, updateTime=1761628596913, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919837569741561, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982527316711, code=themeColor, value=null, createTime=1761628596918, updateTime=1761628596918, creator=18614031015, updator=18614031015), WebsiteProps(id=1189919837573935866, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189918982527316711, code=themeStyle, value=null, createTime=1761628596919, updateTime=1761628596919, creator=18614031015, updator=18614031015)])], journalTitle=汽车工程师, weixinUrl=null, journalUrl=https://tjqc.cbpt.cnki.net/, iacademicId=null, status=1, seqNo=null, journalTitleEn=Automotive Engineer, journalPhotoCn=+bJsKkKt/pjz9u6EwhnksQ==, journalPhotoEn=O3Sn3tnYYrh/jm6emnnMWA==, journalFirstLetter=A, 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), detailUrlCn=https://castjournals.cast.org.cn/joweb/qcgcs/CN/10.20104/j.cnki.1674-6546.20240371, detailUrlEn=https://castjournals.cast.org.cn/joweb/qcgcs/EN/10.20104/j.cnki.1674-6546.20240371, pdfUrlCn=https://castjournals.cast.org.cn/joweb/qcgcs/CN/PDF/10.20104/j.cnki.1674-6546.20240371, pdfUrlEn=https://castjournals.cast.org.cn/joweb/qcgcs/EN/PDF/10.20104/j.cnki.1674-6546.20240371, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于机器学习的汽车门槛梁碰撞仿真优化
收藏切换
PDF下载
刘家员 , 彭竑维 , 辜冠荣
汽车工程师 | 2024年车辆安全与智能交通国际学术会议专题 2025,(6): 41-48
收起
收藏切换
汽车工程师 | 2024年车辆安全与智能交通国际学术会议专题 2025, (6): 41-48
基于机器学习的汽车门槛梁碰撞仿真优化
全屏
刘家员 , 彭竑维, 辜冠荣
作者信息
  • 澳汰尔工程软件(上海)有限公司, 上海 200040

通讯作者:

刘家员(1991—),硕士研究生,工程师,主要研究方向为汽车被动安全,
Crash Simulation Optimization of Automotive Sill Beams Based on Machine Learning
Jiayuan Liu , Hongwei Peng, Guanrong Gu
Affiliations
  • Altair Engineering Software (Shanghai) Co., Ltd., Shanghai 200040
出版时间: 2025-06-15 doi: 10.20104/j.cnki.1674-6546.20240371
文章导航
收藏切换

为提高汽车门槛梁优化效率,针对汽车侧面柱碰撞工况门槛挤压铝吸能量有限、仿真中显式求解迭代周期长及制造工艺要求高等优化难点,基于某SUV车型建立碰撞模型,以门槛梁厚度为变量进行试验设计(DOE)分析,获得144组有效设计数据,通过深度学习方法(Rapidminer、romAI)形成降阶模型(ROM)用作优化仿真的替代模型,CAE仿真验证结果表明,在数据量有限的条件下,romAI在优化中的精度达到95%以上,且求解速度提高近40倍,极大缩短了研发周期。

机器学习  /  降阶模型  /  汽车门槛梁  /  优化仿真

The resrearch aims of the paper is to improve the optimization efficiency of the automobile sill beam, and address the challenges of optimization such as the limited energy absorption of the sill extrusion aluminum under the condition of the side column collision, long iterative cycle of explicit solution in the simulation and the high requirements of the manufacturing process. A collision model is established based on an SUV model, and the Design Of Experiment (DOE) analysis is made with sill beam thickness as variables, resulting in 144 groups of valid design data are obtained. A Reduced-Order Model (ROM) is formed by deep learning methods(Rapidminer, romAI) and used as the alternative model of optimization and simulation. The results of CAE simulation verifies that accuracy of romAI reaches more than 95% in optimization and the solution speed is increased by more than 40 times under the limited data, which greatly shortens the R&D cycle.

Machine learning  /  Reduced Order Model (ROM)  /  Automobile sill beam  /  Optimization simulations
刘家员, 彭竑维, 辜冠荣. 基于机器学习的汽车门槛梁碰撞仿真优化. 汽车工程师, 2025 , (6) : 41 -48 . DOI: 10.20104/j.cnki.1674-6546.20240371
Jiayuan Liu, Hongwei Peng, Guanrong Gu. Crash Simulation Optimization of Automotive Sill Beams Based on Machine Learning[J]. Automotive Engineer, 2025 , (6) : 41 -48 . DOI: 10.20104/j.cnki.1674-6546.20240371
近年来,随着大数据挖掘技术的不断发展,机器学习在汽车仿真领域逐渐开始应用。Oishi等[1]将深度学习引入有限元分析,提出了一种用于网格单元刚度矩阵优化的数值积分方法,相较于传统的数值计算方法,计算速度显著提升。Deshpande等[2]提出了一种基于卷积神经网络的仿真模拟框架,可以实时准确地预测物体承载后的形变。另外,Liang等[3]开发了一种深度学习模型,能够替代有限元分析,快速输出主动脉壁应力分布情况。郭道一等[4-6]基于大数据学习建立了自动驾驶防撞预警算法,可预测汽车碰撞伤害程度。刘淼等[7-12]将大数据分析技术应用于汽车零部件应力分析,实现了汽车结构零件性能预测的突破。早期,刘家员等[13]提出了子模型优化方法,在后续工程应用中得到了广泛采用,然而,CAE与人工智能(Artificial Intelligence,AI)的联合应用进一步减少了工作量,相较于子模型优化方法,效率显著提高。
仿真驱动设计已成为工程研发中的重要手段,其中,优化设计是CAE仿真的核心工作之一。随着技术的不断进步,机器学习降阶模型(Reduced Order Model,ROM)在汽车仿真中的应用日益增加,通过替代部分求解器的工作,提高了后续仿真和优化的效率,为参数化优化的快速迭代提供了新的可能。这种方式在对求解效率要求高的领域应用逐渐增多,如汽车碰撞分析、计算流体力学(Computational Fluid Dynamics,CFD)仿真,以及重型机械颗粒仿真等领域。
引入机器学习技术后,仿真优化过程通常先将变量快速离散化并进行试验设计(Design of Experiment,DOE)分析,然后通过深度学习数据处理构建降阶模型,该模型可直接与HyperStudy等优化工具集成,从而快速找到最优解,大幅提高仿真效率。本文基于实际工程项目,将机器学习应用于汽车侧面柱碰撞工况的门槛梁仿真[14],通过深度学习构建降阶模型并应用于后续优化分析。
侧面柱碰撞工况是中国新车评价规程(China New Car Assessment Programme,C-NCAP)中的必做试验项目,关系到新能源汽车的整车结构安全性[15]。随着整车制造商对电池包防护要求的不断提高,门槛梁的结构优化成为仿真分析的重点研究对象[15-16]。以往,整车门槛加强筋厚度优化仿真普遍采用试错法[17],效率较低。目前,汽车门槛普遍采用挤压铝型材,内部加强筋布置密集且车型间差异显著,同时,优化过程需综合考虑制造工艺约束,计算资源需求大幅增加,尤其是在多变量单一厚度验证场景下。因此,显式求解所需时间成本更高。
本文基于某SUV车型,采用1D梁单元和2D壳单元相结合进行简化建模,可显著提高基础模型的迭代效率,并在保证模型完整性的前提下,细化重点关注的门槛区域,以提高计算精度。利用前处理软件HyperMesh和显式动力学软件Radioss进行建模和求解,简化模型如图1所示。
本文以优化后的侵入量和碰撞挤压力作为评价指标,重点考查侵入量。为保证侵入量测量的准确性,在4个测量点设置侵入量测量弹簧装置,如图2所示。
门槛加强筋厚度范围设定为2.0~3.5 mm;优化响应为门槛的侵入量和挤压力,其中门槛最大侵入量作为约束条件;优化目标为门槛质量最小。
本文基于Radioss开展试验设计分析和求解,获得门槛最大侵入量和门槛挤压力,并根据新能源汽车侧面柱碰撞挤压的特点及项目工程经验,设定目标侵入量为155 mm、目标挤压力为250 kN。基于《C-NCAP管理规则(2024年版)》规定的试验工况,通过后处理分析获得相关响应曲线,如图3所示。从图3中可以看出,门槛最大侵入量为192.3 mm、最大挤压力为236.6 kN,侵入量超过目标要求,故本文以侵入量作为主要优化目标进行分析。
在汽车仿真领域,将深度学习与传统CAE方法相结合,形成了“数据+仿真”的新方法,其流程如图4所示:首先,基于基础模型结果进行变量的试验设计得到训练样本集;然后,利用深度学习构建降阶模型;最后,基于降阶模型进行高效协同优化。
基于已建立的CAE模型,在HyperStudy中以门槛梁厚度为变量开展试验设计分析。DOE求解器共获得144组有效数据,作为后续深度学习的样本数据,如图5所示。
Kurkova等[18-22]对神经网络的泛化能力进行了系统研究,为深度学习的工程应用提供了强有力的理论支持。在模型构建过程中,网络参数的合理设置直接影响结果的可靠性,其中隐藏层节点数量Nh的确定尤为关键,其计算公式为:
${N}_{\mathrm{h}}=\frac{{N}_{s}}{\alpha ({N}_{i}+{N}_{o})}$
式中:NiNo分别为输入层、输出层神经元数量;Ns为训练集样本数量;α为调节系数,取值范围一般为2~10。
通常,隐藏层神经元数量的确定遵循以下原则:在输入层和输出层规模之间;推荐为输入层神经元数量的2/3与输出层神经元数量的2/3之和;应小于输入层神经元数量的2倍[23]
本文采用Knowledge Studio对门槛梁的DOE分析结果进行机器学习处理,将144组原始数据中的120组用于深度学习训练,其余24组用于模型验证,深度学习过程如图6所示。
深度学习模型构建过程通常需要较多的训练数据,本文采用的样本量有限,模型精度有待提高。为提高模型的性能,本文分别通过调整模型的宽度和深度进一步验证其有效性,如表1所示。
表1可知:
a.模型宽度较大时,训练精度较高,但测试精度显著下降,模型出现严重过拟合;模型宽度较小时,训练和测试精度取得平衡,泛化性能最佳;中等模型宽度获得的训练精度较高,但测试精度显著低于训练精度,仍存在过拟合现象。
b.模型深度为1时,模型泛化能力最优,训练与测试精度差距最小;模型深度为2时,测试结果随模型宽度变化波动较大,需结合宽度进行综合优化;模型深度为3时,测试精度显著低于训练精度,泛化能力差。
所以在采用Knowledge Studio进行数据分析时,推荐优先选择模型宽度为3、模型深度为2或模型宽度为10、模型深度为1的参数组合,以兼顾泛化能力与预测精度。此外,应避免使用高宽度(≥50)或高深度(≥3)的配置,以防止过拟合现象。
本文基于机器学习优化结果,与HyperStudy开展联合仿真优化。依托Altair集成化平台的优势,HyperStudy可以直接调用Knowledge Studio训练完成的深度学习模型数据。优化完成后进一步开展CAE仿真验证,精度对比结果如表2所示。
表2可知:全局响应算法(Global Response Surface Method,GRSM)在侵入量优化方面表现较好,但在挤压力优化方面表现一般;序列二次规划(Sequential Quadratic Programming,SQP)在侵入量优化方面同样表现良好,且在挤压力优化方面表现更为均衡,尤其是在模型深度和模型宽度较小的情况下,挤压力差异较小。因此,HyperStudy在挤压力控制方面表现更为均衡,特别是采用SQP方法时。优化方法的选择应根据具体需求:对于以侵入量控制为主要目标的优化问题,建议采用深度学习优化结合GRSM或SQP;需要兼顾侵入量和挤压力时,SQP是更优的选择。
降阶模型是一种具有高保真度的模型简化方法,能够在保留模型基本特性和主导效应的前提下,大幅减少复杂模型的计算时间和存储空间。
romAI通过内部映射建立输入与输出之间的关系网络[6]。这种映射关系的质量由训练样本的质量、数量、超参数的选择以及可用于样本训练的时间决定,其中,超参数包括神经网络隐藏层数量和激活函数等。基于romAI的大数据深度挖掘技术不仅可用于表达静态模型,还可用于表达动态模型,映射关系如图7所示。
静态模型可表达为:
y=g(u(t),θ)
动态模型可表达为:
$\dot{x}=f\left(x\right(t),u(t),\theta )$
y=g(x(t),u(t),θ)
式中:x为变量,gf为复杂系统的表征函数,uy分别为系统的输入、输出端口,t为系统时间,θ为特征参数。
在数学上,静态系统由代数方程描述,动态系统由微分方程描述,静态方程和动态系统的降阶模型可以是线性的或非线性的,具体取决于未知函数在方程中的表示方式。2012年,Raveh[22]成功将降阶模型应用于汽车空气动力学研究,取得了较好的工程实践效果。
romAI采用前馈神经网络(Feedforward Neural Network,FFNN)构建降阶模型,该网络包含多个隐藏层,每个隐藏层由若干个神经元组成,各层神经元接收来自前一层的输出,经过加权和处理后传递至下一层。寇家庆等[24]研究表明,模型宽度对模型精度具有重要影响,特别是在径向基函数(Radial Basis Function,RBF)神经网络气动力建模中,模型宽度的选择会显著改变模型的泛化能力。
本文基于现有门槛梁CAE仿真数据建立相应的降阶模型。romAI的输入数据仍为整车门槛梁设计变量,通过Compose可视化编程软件进行脚本处理,再导入120组CSV格式的数据文件对深度学习参数进行控制,并通过预设输入、输出路径实现自动化建模,如图8所示。
对降阶模型进行精度分析,基于参数化深度学习方法,采用曲线匹配度作为评价指标,匹配度目标设为90%。训练模型的预测曲线与基准曲线展现出良好的吻合特性,如图9所示。
在前处理模型设置中,除配置模型的输入、输出路径及名称外,还要设置模型深度、模型宽度和模型计算迭代步数,以及划分为测试集的样本数量。基于romAI构建的降阶模型如图10所示。
HyperStudy是平台化的参数优化工具,可与多种有限元分析求解器连接,实现从试验设计到优化的全过程分析。HyperStudy直接调用romAI构建的高保真降阶模型,优化结果可通过CAE仿真进行验证。
降阶模型可以转换为HyperStudy可识别的CSV格式,完成门槛梁厚度优化。优化后,将HyperStudy输出的门槛梁厚度最终优化方案带入原始CAE模型中验证即可。优化结果与实际制造工艺要求存在差异时,可根据工程经验进行适当调整,如表3所示,其中加强筋的相应位置见图1
HyperStudy提供了多种优化方法,包括自适应响应面法(Adaptive Response Surface Methodology,ARSM)、遗传算法(Genetic Algorithm,GA)、GRSM等,本文选择GRSM。GRSM是基于响应面的直接优化方法,尤其在多目标参数优化问题中展现出显著的计算效率和工程实用性优势。
romAI优化结果与CAE分析结果如表4所示,其误差控制在5%以内,表明降阶模型的预测结果与实际仿真结果具有较高的一致性。验证结果表明,针对研发中的疑难问题,可以采用该方法辅助快速评估设计方案,以缩短研发周期。
在传统的有限元仿真中,对比分析是仿真精度可靠性分析的重要一环。Kou等[25-26]的研究表明,降阶模型在工程应用中具有较高的预测精度。本文方案以传统有限元仿真为基础,可靠性较高,采用2种深度学习工具对汽车门槛梁厚度优化结果进行对比,最后通过CAE仿真[8]验证其可靠性,结果如表5所示。
表5可以看出,romAI在降阶模型生成方面精度更高,其构建的降阶模型在后续优化应用中误差低于5%。此外,在时效性方面,本文使用标准的1D DEMO模型求解器求解耗时665 s,采用降阶模型仅需17 s左右即可完成。机器学习方法使用约1/40的时间成本实现了95%的精度。如需进一步提高精度,需要增加训练样本量[6]
针对本文的SUV车型进一步分析romAI的优化结果,不同时刻的侵入量和碰撞挤压力曲线[7]显示出良好的一致性。门槛梁的柱碰撞侵入量输出结果对比曲线如图11所示,碰撞过程前期,机器学习模型的预测精度接近100%,碰撞后期精度轻微下降,整体仍保持较高的吻合度。
本文基于整车侧面柱碰撞工况的门槛梁CAE仿真,通过2次数据深度学习实现了门槛梁厚度优化。结果表明,基于降阶模型的深度学习优化方法效率较高,特别是romAI降阶模型,在样本数量有限的条件下,可保证较高的精度并缩短研发周期。
机器学习优化结果与CAE仿真结果具有较高的一致性,可以作为前期开发的有效预测手段。Altair软件平台的集成应用实现了降阶模型的直接调用和数据高效传递,显著提高了研发效率。
参考文献 引证文献
排序方式:
[1]
OISHI A, YAGAWA G. Computational Mechanics Enhanced by Deep Learning[J]. Computer Methods in Applied Mechanics and Engineering, 2017, 327: 327-351.
[2]
DESHPANDE S, LENGIEWICZ J, BORDAS S. Probabilistic Deep Learning for Real-Time Large Deformation Simulations[J]. Computer Methods in Applied Mechanics and Engineering, 2022, 398.
[3]
LIANG L, LIU M L, MARTIN C, et al. A Deep Learning Approach to Estimate Stress Distribution: A Fast and Accurate Surrogate of Finite-Element Analysis[J]. Journal of the Royal Society Interface, 2018, 15(138).
[4]
郭道一. 基于深度学习和仿真的汽车碰撞损伤预测方法[D]. 大连: 大连理工大学, 2019.
GUO D Y. Automobile Crash Damage Prediction Method Based on Deep Learning and Simulation[D]. Dalian: Dalian University of Technology, 2019.
[5]
祝雪峰, 傅文谦, 郭道一. 一种基于仿真数据和人工智能的汽车碰撞损伤等级实时预测方法: CN113094808A[P]. 2021-07-09.
ZHU X F, FU W Q, GUO D Y. A Real-Time Prediction Method for Automobile Crash Damage Grade Based on Simulation Data and Artificial Intelligence: CN113094808A[P]. 2021-07-09.
[6]
高军. 基于计算机视觉与深度学习的汽车主动安全系统关键技术研究[D]. 武汉: 武汉理工大学, 2018.
GAO J. Research on Key Technologies of Automotive Active Safety System Based on Computer Vision and Deep Learning[D]. Wuhan: Wuhan University of Technology, 2018.
[7]
刘淼, 孙成智, 王潇嵩. 基于深度学习的汽车结构件应力预测方法[C]// 2021中国汽车工程学会年会论文集. 上海: 机械工业出版社, 2021: 591-594.
LIU M, SUN C Z, WANG X S. Stress Prediction Method for Automotive Structural Parts Based on Deep Learning[C]// Proceedings of the 2021 Annual Conference of the Society of Automotive Engineers of China. Shanghai: China Machine Press, 2021: 591-594.
[8]
高泽, 楚遵康, 石稼晟, 等. 基于图网络的汽车零部件应力场快速预测方法研究[J]. 汽车工程, 2024, 46(1): 170-178.
GAO Z, CHU Z K, SHI J S, et al. Research on Rapid Prediction Method of Stress Field of Auto Parts Based on Graph Network[J]. Automotive Engineering, 2024, 46(1): 170-178.
[9]
王恩铭. 基于深度学习与有限元方法的细观泡沫金属压缩力学性能研究[D]. 长春: 吉林大学, 2024.
WANG E M. Study on Mechanical Properties of Mesoscopic Metal Foam Based on Deep Learning and Finite Element Method[D]. Changchun: Jilin University, 2024.
[10]
杨静, 张会, 冯旸赫, 等. 基于深度学习的作战仿真时序数据处理与分类问题研究[C]// 第八届中国指挥控制大会. 北京: 中国指挥与控制学会, 2020: 617-623.
YANG J, ZHANG H, FENG Y H, et al. Research on Time Series Data Processing and Classification of Combat Simulation Based on Deep Learning[C]// The 8th China Command and Control Conference. Beijing: Chinese Command and Control Society, 2020: 617-623.
[11]
杨叶倩. 基于深度学习的碳纤维制备工艺双向优化与仿真算法研究[D]. 济南: 山东大学, 2023.
YANG Y Q. Research on Two-Way Optimization and Simulation Algorithm of Carbon Fiber Preparation Process Based on Deep Learning[D]. Jinan: Shandong University, 2023.
[12]
黄泽辉, 唐洪斌, 王雪松, 等. 基于深度学习的车辆碰撞乘员伤害预测方法研究[J]. 汽车工程师, 2024(4): 8-11.
HUANG Z H, TANG H B, WANG X S, et al. Research on Vehicle Collision Occupant Injury Prediction Method Based on Deep Learning[J]. Automotive Engineer, 2024(4): 8-11.
[13]
刘家员, 冷川. 基于汽车架构子模型的快速仿真方法[J]. 计算机辅助工程, 2024, 33(4): 21-24.
LIU J Y, LENG C. Fast Simulation Method Based on Submodel of Vehicle Architecture[J]. Computer Aided Engineering, 2024, 33(4): 21-24.
[14]
唐波, 赵晓红. 汽车侧面柱碰撞的研究以及法规动态[J]. 汽车与配件, 2012(49): 24-26.
TANG B, ZHAO X H. Research on Side Pillar Collision of Automobile and Regulatory Dynamics[J]. Auto & Accessories, 2012(49): 24-26.
[15]
范亚彬. 汽车侧面柱碰撞试验装置设计及仿真优化[D]. 天津: 河北工业大学, 2017.
FAN Y B. Design and Simulation Optimization of Automobile Side Column Crash Test Device[D]. Tianjin: Hebei University of Technology, 2017.
[16]
李凯. 纯电动汽车动力锂电池箱底部碰撞防护结构设计与仿真分析[D]. 长沙: 中南大学, 2023.
LI K. Design and Simulation Analysis of Collision Protection Structure at the Bottom of Power Lithium Battery Box of Pure Electric Vehicle[D]. Changsha: Central South University, 2023.
[17]
朱啟兵, 吴业波, 曹雄, 等. 防护性强的电动车锂电池用保护板支架: CN216488291U[P]. 2022-05-10.
ZHU Q B, WU Y B, CAO X, et al. Protection Plate Bracket for Lithium Battery of Electric Vehicle with Strong Protection: CN216488291U[P]. 2022-05-10.
[18]
KURKOVA V, SANGUINETI M. Learning with Generalization Capability by Kernel Methods of Bounded Complexity[J]. Journal of Complexity, 2005, 21(3): 350-367.
[19]
ZHAO Z Q, HUANG D S. A Mended Hybrid Learning Algorithm for Radial Basis Function Neural Networks to Improve Generalization Capability[J]. Applied Mathematical Modelling, 2007, 31(7): 1271-1281.
[20]
LEUNG C T, CHOW T W S. Adaptive Regularization Parameter Selection Method for Enhancing Generalization Capability of Neural Networks[J]. Artificial Intelligence, 1999, 107(2): 347-356.
[21]
HALL K C, THOMAS J P, DOWELL E H. Proper Orthogonal Decomposition Technique for Transonic Unsteady Aerodynamic Flows[J]. AIAA Journal, 2000, 38(10): 1853-1862.
[22]
RAVEH D E. Reduced-Order Models for Nonlinear Unsteady Aerodynamics[J]. AIAA Journal, 2012, 39(8): 1417-1429.
[23]
BENGIO Y, COURVILLE A, VINCENT P. Representation Learning: A Review and New Perspectives[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2013, 35(8): 1798-1828.
[24]
寇家庆, 张伟伟. 基函数宽度对递归RBF神经网络气动力模型精度的影响研究[J]. 航空工程进展, 2015, 6(3): 261-270.
KOU J Q, ZHANG W W. Research on the Effects of Basis Function Widths of Aerodynamic Modeling Based on Recursive RBF Neural Network[J]. Advancesin Aeronautical Scienceand Engineering, 2015, 6(3): 261-270.
[25]
GLAZ B, LIU L, FRIEDMANN P P. Reduced-Order Nonlinear Unsteady Aerodynamic Modeling Using a Surrogate-Based Recurrence Framework[J]. AIAA Journal, 2012, 48(10): 2418-2429.
[26]
STORN R, PRICE K. Differential Evolution-A Simple and Efficient Heuristic for Global Optimization over Continuous Spaces[J]. Journal of Global Optimization, 1997, 11(4): 341-359.
2025年第卷第6期
PDF下载
195
80
引用本文
BibTeX
文章信息
doi: 10.20104/j.cnki.1674-6546.20240371
  • 首发时间:2025-11-10
  • 出版时间:2025-06-15
补充材料
相关文章
文章信息
作者
出版历史
  • 修回日期:2025-02-11
基金
作者信息
    澳汰尔工程软件(上海)有限公司, 上海 200040

通讯作者:

刘家员(1991—),硕士研究生,工程师,主要研究方向为汽车被动安全,
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/qcgcs/CN/10.20104/j.cnki.1674-6546.20240371
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

Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)

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