Article(id=1263514362317254680, tenantId=1146029695717560320, journalId=1263187241531621409, issueId=1263514351571428296, articleNumber=null, orderNo=null, doi=10.11996/JG.j.2095-302X.2026010173, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=research-article, receivedDate=1744473600000, receivedDateStr=2025-04-13, revisedDate=null, revisedDateStr=null, acceptedDate=1754582400000, acceptedDateStr=2025-08-08, onlineDate=1779174898938, onlineDateStr=2026-05-19, pubDate=1772208000000, pubDateStr=2026-02-28, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1779174898938, onlineIssueDateStr=2026-05-19, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1779174898938, creator=13701087609, updateTime=1779174898938, updator=13701087609, issue=Issue{id=1263514351571428296, tenantId=1146029695717560320, journalId=1263187241531621409, year='2026', volume='47', issue='1', pageStart='1', pageEnd='233', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=1, specialIssue=null, createTime=1779174896376, creator=13701087609, updateTime=1779174963943, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1263514635077039012, tenantId=1146029695717560320, journalId=1263187241531621409, issueId=1263514351571428296, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1263514635077039013, tenantId=1146029695717560320, journalId=1263187241531621409, issueId=1263514351571428296, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=173, endPage=178, ext={EN=ArticleExt(id=1263514364502487080, articleId=1263514362317254680, tenantId=1146029695717560320, journalId=1263187241531621409, language=EN, title=Generative digital twin modeling based on large models, columnId=1263514353999930314, journalTitle=Journal of Graphics, columnName=Digital Design and Manufacture, runingTitle=null, highlight=null, articleAbstract=

To address the challenges in integrating Digital-Twin (DT) technology with large-scale generative models in industrial design, a CAD-LDT digital-twin modeling framework based on generative foundation models was proposed. The framework adopted a triadic architecture consisting of a physical-entity module, an intelligent generation module, and a virtual-entity module, and innovatively incorporated multi-modal data fusion mechanisms and domain-knowledge constraints to enable autonomous generation of parameterized CAD models from physical-entity descriptions. Utilizing LLaVA-7B and LLaMA-7B as backbone models, the framework employed LoRA-based lightweight adapters to achieve cross-modal alignment between visual and textual features, and introduced a constraint encoder that transformed geometric tolerances and physical rules into structured JSON objects. To enhance the mathematical consistency of spatial transformations, Lie-group algorithms were adopted for the optimization of rigid-body transformations, while a geometric-weight binning strategy was proposed to discretize complex assembly relationships. Moreover, a spatiotemporal-decoupled generation strategy was designed to jointly optimize spatial layout and assembly sequencing. Experimental results on the DeepCAD dataset indicated that the proposed framework achieved an Intersection- over-Union (IoU) of 83.6%, a constraint satisfaction rate of 91.3%, and a 26.5% improvement in generation efficiency, significantly outperforming existing baseline models. Further ablation studies confirmed the critical contributions of multi-modal fusion, constraint encoding mechanisms, and Lie-group optimization to modeling performance, providing a novel DT modeling paradigm for intelligent manufacturing with demonstrated value in parametric design and assembly process optimization.

, correspAuthors=null, authorNote=null, correspAuthorsNote=
LIANG Shenglong,E-mail:
, 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=Shenglong LIANG, Qiuxia FAN), CN=ArticleExt(id=1263514373847396490, articleId=1263514362317254680, tenantId=1146029695717560320, journalId=1263187241531621409, language=CN, title=基于大模型的生成式数字孪生体建模, columnId=1263514354243199948, journalTitle=图学学报, columnName=数字化设计与制造, runingTitle=null, highlight=null, articleAbstract=

针对数字孪生(DT)技术与大模型在工业设计中融合应用的挑战,提出了一种基于生成式大模型的CAD-LDT数字孪生体建模框架。通过构建物理实体模块、智能生成模块和虚拟实体模块的三元架构,创新性地整合多模态数据融合机制与领域知识约束,实现从物理实体描述到参数化CAD模型的自主生成。采用LLaVA-7B和LLaMA-7B作为基础模型,通过LoRA轻量化适配器实现视觉-文本特征的跨模态对齐,并设计约束编码器将几何公差与物理规则转化为结构化JSON对象。为增强空间变换的数学一致性,引入李群算法优化刚体变换表征;采用几何权重分箱方法离散化复杂装配关系;提出时空解耦生成策略,协同规划空间布局与装配时序。在DeepCAD数据集上的实验结果表明,该框架在几何达到83.6%,约束满足率达91.3%,生成效率提升了26.5%,显著优于主流基线模型。消融实验进一步验证了多模态融合、约束编码机制和李群算法对建模质量的关键贡献。为智能制造领域提供了新的DT建模范式,在参数化设计、装配工艺优化等方面展现出工程应用价值。

, correspAuthors=null, authorNote=null, correspAuthorsNote=
梁生龙,E-mail:
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=nC+nM56BPljN4nZEsgd/EA==, magXml=sNonygZZATzOHDKIy+NHOw==, pdfUrl=null, pdf=XW+DF736g5nJAc81Z6LNUQ==, pdfFileSize=2608571, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=QaPp3XnXv0AZDAxA6kPGjg==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=ZDPqlztXhgGXAPn/cf+Ayw==, mapNumber=null, authorCompany=null, fund=null, authors=null, authorsList=梁生龙, 范秋霞)}, authors=[Author(id=1263550807958536800, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=a15919281689@126.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1263550808055005796, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, authorId=1263550807958536800, language=EN, stringName=Shenglong LIANG, firstName=Shenglong, middleName=null, lastName=LIANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 College of Mechanical and Electrical Engineering, Zhuhai City Vocational and Technical College, Zhuhai Guangdong 519090, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1263550808440881766, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, authorId=1263550807958536800, language=CN, stringName=梁生龙, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 珠海城市职业技术学院机电工程学院, 广东 珠海 519090, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1263550806947709520, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, xref=1, ext=[AuthorCompanyExt(id=1263550807098704465, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, companyId=1263550806947709520, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 College of Mechanical and Electrical Engineering, Zhuhai City Vocational and Technical College, Zhuhai Guangdong 519090, China), AuthorCompanyExt(id=1263550807186784850, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, companyId=1263550806947709520, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 珠海城市职业技术学院机电工程学院, 广东 珠海 519090)])]), Author(id=1263550808784814701, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, 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=1263550809204245105, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, authorId=1263550808784814701, language=EN, stringName=Qiuxia FAN, firstName=Qiuxia, middleName=null, lastName=FAN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 School of Automation and Software, Shanxi University, Taiyuan Shanxi 030006, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1263550810911326841, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, authorId=1263550808784814701, language=CN, stringName=范秋霞, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 山西大学自动化与软件学院, 山西 太原 030006, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1263550807555883609, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, xref=2, ext=[AuthorCompanyExt(id=1263550807610409562, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, companyId=1263550807555883609, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 School of Automation and Software, Shanxi University, Taiyuan Shanxi 030006, China), AuthorCompanyExt(id=1263550807622992475, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, companyId=1263550807555883609, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 山西大学自动化与软件学院, 山西 太原 030006)])])], keywords=[Keyword(id=1263550811745993354, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, language=EN, orderNo=1, keyword=large models), Keyword(id=1263550812198978193, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, language=EN, orderNo=2, keyword=digital twin), Keyword(id=1263550812601631382, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, language=EN, orderNo=3, keyword=multimodal data), Keyword(id=1263550813436297883, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, language=EN, orderNo=4, keyword=intelligent manufacturing), Keyword(id=1263550813935420065, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, language=EN, orderNo=5, keyword=parametric design), Keyword(id=1263550815474729640, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, language=CN, orderNo=1, keyword=大模型), Keyword(id=1263550816309396143, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, language=CN, orderNo=2, keyword=数字孪生), Keyword(id=1263550816791741112, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, language=CN, orderNo=3, keyword=多模态数据), Keyword(id=1263550817655767744, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, language=CN, orderNo=4, keyword=智能制造), Keyword(id=1263550818461074118, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, language=CN, orderNo=5, keyword=参数化设计)], refs=[Reference(id=1263550847540183983, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=4, pageStart=43, pageEnd=50, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=王剑, 王好臣, 李学伟, journalName=现代制造工程, refType=null, unstructuredReference=王剑, 王好臣, 李学伟, . 基于OPC UA的数字孪生车间信息物理融合系统[J]. 现代制造工程, 2023(4): 43-50., articleTitle=基于OPC UA的数字孪生车间信息物理融合系统, refAbstract=null), Reference(id=1263550848240632754, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=4, pageStart=43, pageEnd=50, url=null, language=null, rfNumber=[1], rfOrder=1, authorNames=WANG J, WANG H C, LI X W, journalName=Modern Manufacturing Engineering, refType=null, unstructuredReference=WANG J, WANG H C, LI X W, et al. Digital twin workshop information physical fusion system based on OPC UA[J]. Modern Manufacturing Engineering, 2023(4): 43-50 (in Chinese)., articleTitle=Digital twin workshop information physical fusion system based on OPC UA, refAbstract=null), Reference(id=1263550849511506872, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=null, pageStart=85, pageEnd=null, url=null, language=null, rfNumber=[2], rfOrder=2, authorNames=GRIEVES M, VICKERS J, KAHLEN F J, FLUMERFELT S, ALVES A, journalName=Transdisciplinary Perspectives on Complex Systems:New Findings and Approaches, refType=null, unstructuredReference=GRIEVES M, VICKERS J. Digital twin: mitigating unpredictable, undesirable emergent behavior in complex systems[M]// KAHLEN F J, FLUMERFELT S, ALVES A. Transdisciplinary Perspectives on Complex Systems:New Findings and Approaches. Cham: Springer, 2017: 85-113., articleTitle=null, refAbstract=null), Reference(id=1263550850392310722, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, doi=null, pmid=null, pmcid=null, year=2023, volume=21, issue=2, pageStart=112, pageEnd=116, url=null, language=null, rfNumber=[3], rfOrder=3, authorNames=王进峰, 问丛川, 花广如, journalName=中国工程机械学报, refType=null, unstructuredReference=王进峰, 问丛川, 花广如. 面向概念、技术与应用的数字孪生综述[J]. 中国工程机械学报, 2023, 21(2): 112-116, 133., articleTitle=面向概念、技术与应用的数字孪生综述, refAbstract=null), Reference(id=1263550852392993735, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, doi=null, pmid=null, pmcid=null, year=2023, volume=21, issue=2, pageStart=112, pageEnd=116, url=null, language=null, rfNumber=[3], rfOrder=4, authorNames=WANG J F, WEN C C, HUA G R, journalName=Chinese Journal of Construction Machinery, refType=null, unstructuredReference=WANG J F, WEN C C, HUA G R. A survey of digital twins for concept, technology and application[J]. Chinese Journal of Construction Machinery, 2023, 21(2): 112-116, 133 (in Chinese)., articleTitle=A survey of digital twins for concept, technology and application, refAbstract=null), Reference(id=1263550852904698827, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, doi=null, pmid=null, pmcid=null, year=2013, volume=46, issue=7, pageStart=150, pageEnd=156, url=null, language=null, rfNumber=[4], rfOrder=5, authorNames=LEE J, LAPIRA E, YANG S H, journalName=IFAC Proceedings Volumes, refType=null, unstructuredReference=LEE J, LAPIRA E, YANG S H, et al. Predictive manufacturing system-trends of next-generation production systems[J]. IFAC Proceedings Volumes, 2013, 46(7): 150-156., articleTitle=Predictive manufacturing system-trends of next-generation production systems, refAbstract=null), Reference(id=1263550853768725457, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, doi=null, pmid=null, pmcid=null, year=2019, volume=13, issue=4, pageStart=534, pageEnd=556, url=null, language=null, rfNumber=[5], rfOrder=6, authorNames=BAO J S, GUO D S, LI J, journalName=Enterprise Information Systems, refType=null, unstructuredReference=BAO J S, GUO D S, LI J, et al. The modelling and operations for the digital twin in the context of manufacturing[J]. Enterprise Information Systems, 2019, 13(4): 534-556., articleTitle=The modelling and operations for the digital twin in the context of manufacturing, refAbstract=null), Reference(id=1263550856184644569, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, doi=null, pmid=null, pmcid=null, year=2018, volume=96, issue=1/4, pageStart=1149, pageEnd=1163, url=null, language=null, rfNumber=[6], rfOrder=7, authorNames=ZHUANG C B, LIU J H, XIONG H, journalName=The International Journal of Advanced Manufacturing Technology, refType=null, unstructuredReference=ZHUANG C B, LIU J H, XIONG H. Digital twin-based smart production management and control framework for the complex product assembly shop-floor[J]. The International Journal of Advanced Manufacturing Technology, 2018, 96(1/4): 1149-1163., articleTitle=Digital twin-based smart production management and control framework for the complex product assembly shop-floor, refAbstract=null), Reference(id=1263550857023505373, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, doi=null, pmid=null, pmcid=null, year=2018, volume=67, issue=1, pageStart=169, pageEnd=172, url=null, language=null, rfNumber=[7], rfOrder=8, authorNames=TAO F, ZHANG M, LIU Y S, journalName=CIRP Annals, refType=null, unstructuredReference=TAO F, ZHANG M, LIU Y S, et al. Digital twin driven prognostics and health management for complex equipment[J]. CIRP Annals, 2018, 67(1): 169-172., articleTitle=Digital twin driven prognostics and health management for complex equipment, refAbstract=null), Reference(id=1263550857887532002, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, doi=null, pmid=null, pmcid=null, year=2022, volume=118, issue=1/2, pageStart=551, pageEnd=570, url=null, language=null, rfNumber=[8], rfOrder=9, authorNames=JIANG Y K, CHEN J H, ZHOU H C, journalName=The International Journal of Advanced Manufacturing Technology, refType=null, unstructuredReference=JIANG Y K, CHEN J H, ZHOU H C, et al. Contour error modeling and compensation of CNC machining based on deep learning and reinforcement learning[J]. The International Journal of Advanced Manufacturing Technology, 2022, 118(1/2): 551-570., articleTitle=Contour error modeling and compensation of CNC machining based on deep learning and reinforcement learning, refAbstract=null), Reference(id=1263550858701226986, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, doi=null, pmid=null, pmcid=null, year=2023, volume=17, issue=6, pageStart=847, pageEnd=859, url=null, language=null, rfNumber=[9], rfOrder=10, authorNames=SAMSONOV V, CHRISMARIE E, KÖPKEN H G, journalName=Production Engineering, refType=null, unstructuredReference=SAMSONOV V, CHRISMARIE E, KÖPKEN H G, et al. Deep representation learning and reinforcement learning for workpiece setup optimization in CNC milling[J]. Production Engineering, 2023, 17(6): 847-859., articleTitle=Deep representation learning and reinforcement learning for workpiece setup optimization in CNC milling, refAbstract=null), Reference(id=1263550860597052400, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, doi=null, pmid=null, pmcid=null, year=2020, volume=61, issue=null, pageStart=101847, pageEnd=null, url=null, language=null, rfNumber=[10], rfOrder=11, authorNames=LI B R, ZHANG H, YE P Q, journalName=Robotics and Computer-Integrated Manufacturing, refType=null, unstructuredReference=LI B R, ZHANG H, YE P Q, et al. Trajectory smoothing method using reinforcement learning for computer numerical control machine tools[J]. Robotics and Computer-Integrated Manufacturing, 2020, 61: 101847., articleTitle=Trajectory smoothing method using reinforcement learning for computer numerical control machine tools, refAbstract=null), Reference(id=1263550861276529653, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=null, pageStart=8736, pageEnd=8754, url=null, language=null, rfNumber=[11], rfOrder=12, authorNames=YAO S Y, RAO R H, HAUSKNECHT M, journalName=2020 Conference on Empirical Methods in Natural Language Processing, refType=null, unstructuredReference=YAO S Y, RAO R H, HAUSKNECHT M, et al. Keep CALM and explore: language models for action generation in text-based games[C]// 2020 Conference on Empirical Methods in Natural Language Processing. Stroudsburg: ACL, 2020: 8736-8754., articleTitle=Keep CALM and explore: language models for action generation in text-based games, refAbstract=null), Reference(id=1263550861616268286, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=2024, pageEnd=04, url=https://arxiv.org/pdf/2302.09368, language=null, rfNumber=[12], rfOrder=13, authorNames=PANG J C, YANG X Y, YANG S H, journalName=null, refType=null, unstructuredReference=PANG J C, YANG X Y, YANG S H, et al. Natural language-conditioned reinforcement learning with inside-out task language development and translation[EB/OL]. [2024- 04-10]. https://arxiv.org/pdf/2302.09368., articleTitle=Natural language-conditioned reinforcement learning with inside-out task language development and translation, refAbstract=null), Reference(id=1263550862253801477, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, doi=null, pmid=null, pmcid=null, year=2025, volume=31, issue=3, pageStart=760, pageEnd=777, url=null, language=null, rfNumber=[13], rfOrder=14, authorNames=何援军, 于海燕, 方志刚, journalName=计算机集成制造系统, refType=null, unstructuredReference=何援军, 于海燕, 方志刚. 基于形计算重构CAD的计算基础[J]. 计算机集成制造系统, 2025, 31(3): 760-777., articleTitle=基于形计算重构CAD的计算基础, refAbstract=null), Reference(id=1263550862648066059, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, doi=null, pmid=null, pmcid=null, year=2025, volume=31, issue=3, pageStart=760, pageEnd=777, url=null, language=null, rfNumber=[13], rfOrder=15, authorNames=HE Y J, YU H Y, FANG Z G, journalName=Computer Integrated Manufacturing Systems, refType=null, unstructuredReference=HE Y J, YU H Y, FANG Z G. Reconstructing computational foundation for CAD based on shape computing[J]. Computer Integrated Manufacturing Systems, 2025, 31(3): 760-777 (in Chinese)., articleTitle=Reconstructing computational foundation for CAD based on shape computing, refAbstract=null), Reference(id=1263550864803938322, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, doi=null, pmid=null, pmcid=null, year=2025, volume=37, issue=1, pageStart=8, pageEnd=14, url=null, language=null, rfNumber=[14], rfOrder=16, authorNames=黄川, 李雅琴, 祁越然, journalName=自然资源遥感, refType=null, unstructuredReference=黄川, 李雅琴, 祁越然, . 基于3D-CAE的高光谱解混及小样本分类方法[J]. 自然资源遥感, 2025, 37(1): 8-14., articleTitle=基于3D-CAE的高光谱解混及小样本分类方法, refAbstract=null), Reference(id=1263550865324032025, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, doi=null, pmid=null, pmcid=null, year=2025, volume=37, issue=1, pageStart=8, pageEnd=14, url=null, language=null, rfNumber=[14], rfOrder=17, authorNames=HUANG C, LI Y Q, QI Y R, journalName=Remote Sensing for Natural Resources, refType=null, unstructuredReference=HUANG C, LI Y Q, QI Y R, et al. A hyperspectral unmixing and few-shot classification method based on 3DCAE network[J]. Remote Sensing for Natural Resources, 2025, 37(1): 8-14 (in Chinese)., articleTitle=A hyperspectral unmixing and few-shot classification method based on 3DCAE network, refAbstract=null), Reference(id=1263550865714102305, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, doi=null, pmid=null, pmcid=null, year=2025, volume=42, issue=7, pageStart=1964, pageEnd=1971, url=null, language=null, rfNumber=[15], rfOrder=18, authorNames=龙禹辰, 勾智楠, 陈宇欣, journalName=计算机应用研究, refType=null, unstructuredReference=龙禹辰, 勾智楠, 陈宇欣, . 基于大语言模型的多任务生成式重构对话情绪识别[J]. 计算机应用研究, 2025, 42(7): 1964-1971., articleTitle=基于大语言模型的多任务生成式重构对话情绪识别, refAbstract=null), Reference(id=1263550866133532709, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, doi=null, pmid=null, pmcid=null, year=2025, volume=42, issue=7, pageStart=1964, pageEnd=1971, url=null, language=null, rfNumber=[15], rfOrder=19, authorNames=LONG Y C, GOU Z N, CHEN Y X, journalName=Application Research of Computers, refType=null, unstructuredReference=LONG Y C, GOU Z N, CHEN Y X, et al. Multi-task generative emotion recognition in conversation based on large language models[J]. Application Research of Computers, 2025, 42(7): 1964-1971 (in Chinese)., articleTitle=Multi-task generative emotion recognition in conversation based on large language models, refAbstract=null), Reference(id=1263550866787844140, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=2025, pageEnd=04, url=https://github.com/CadQuery/cadquery, language=null, rfNumber=[16], rfOrder=20, authorNames=URBANCZYK A, WRIGHT J, journalName=null, refType=null, unstructuredReference=URBANCZYK A, WRIGHT J. CadQuery: a python parametric CAD scripting framework based on OCCT[EB/OL]. [2025- 04-10]. https://github.com/CadQuery/cadquery., articleTitle=CadQuery: a python parametric CAD scripting framework based on OCCT, refAbstract=null), Reference(id=1263550867211468852, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, doi=null, pmid=null, pmcid=null, year=2024, volume=38, issue=7, pageStart=115, pageEnd=126, url=null, language=null, rfNumber=[17], rfOrder=21, authorNames=刘静文, 刘渊, 袁琮淇, journalName=中文信息学报, refType=null, unstructuredReference=刘静文, 刘渊, 袁琮淇. 融合超图卷积和自监督协同训练的组推荐算法[J]. 中文信息学报, 2024, 38(7): 115-126, 136., articleTitle=融合超图卷积和自监督协同训练的组推荐算法, refAbstract=null), Reference(id=1263550869241511995, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, doi=null, pmid=null, pmcid=null, year=2024, volume=38, issue=7, pageStart=115, pageEnd=126, url=null, language=null, rfNumber=[17], rfOrder=22, authorNames=LIU J W, LIU Y, YUAN C Q, journalName=Journal of Chinese Information Processing, refType=null, unstructuredReference=LIU J W, LIU Y, YUAN C Q. Group recommendation algorithms incorporating hypergraph convolution and self-supervised collaborative training[J]. Journal of Chinese Information Processing, 2024, 38(7): 115-126, 136 (in Chinese)., articleTitle=Group recommendation algorithms incorporating hypergraph convolution and self-supervised collaborative training, refAbstract=null), Reference(id=1263550870000681027, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=http://kns.cnki.net/kcms/detail/10.1034.T.20250331.1207.002.html, language=null, rfNumber=[18], rfOrder=23, authorNames=雷松林, 赵征鹏, 阳秋霞, journalName=null, refType=null, unstructuredReference=雷松林, 赵征鹏, 阳秋霞, . 基于可解耦扩散模型的零样本风格迁移[EB/OL]. (2025-03-31) [2025-04-02]. http://kns.cnki.net/kcms/detail/10.1034.T.20250331.1207.002.html., articleTitle=基于可解耦扩散模型的零样本风格迁移, refAbstract=null), Reference(id=1263550870315253834, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=http://kns.cnki.net/kcms/detail/10.1034.T.20250331.1207.002.html, language=null, rfNumber=[18], rfOrder=24, authorNames=LEI S L, ZHAO Z P, YANG Q X, journalName=null, refType=null, unstructuredReference=LEI S L, ZHAO Z P, YANG Q X, et al. Zero-shot style transfer based on decoupled diffusion models[EB/OL]. (2025-03-31) [2025-04-02]. http://kns.cnki.net/kcms/detail/10.1034.T.20250331.1207.002.html in Chinese)., articleTitle=Zero-shot style transfer based on decoupled diffusion models, refAbstract=null), Reference(id=1263550870545940561, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, doi=null, pmid=null, pmcid=null, year=2024, volume=35, issue=10, pageStart=4806, pageEnd=4825, url=null, language=null, rfNumber=[19], rfOrder=25, authorNames=孙灏铖, 刘力, 李凡长, journalName=软件学报, refType=null, unstructuredReference=孙灏铖, 刘力, 李凡长. 李群模糊C均值聚类图像分割算法[J]. 软件学报, 2024, 35(10): 4806-4825., articleTitle=李群模糊C均值聚类图像分割算法, refAbstract=null), Reference(id=1263550870768238680, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, doi=null, pmid=null, pmcid=null, year=2024, volume=35, issue=10, pageStart=4806, pageEnd=4825, url=null, language=null, rfNumber=[19], rfOrder=26, authorNames=SUN H C, LIU L, LI F Z, journalName=Journal of Software, refType=null, unstructuredReference=SUN H C, LIU L, LI F Z. Lie group fuzzy C-means clustering algorithm for image segmentation[J]. Journal of Software, 2024, 35(10): 4806-4825 (in Chinese)., articleTitle=Lie group fuzzy C-means clustering algorithm for image segmentation, refAbstract=null), Reference(id=1263550870948593756, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, doi=null, pmid=null, pmcid=null, year=2006, volume=null, issue=1, pageStart=32, pageEnd=35, url=null, language=null, rfNumber=[20], rfOrder=27, authorNames=王乃成, journalName=机械工业标准化与质量, refType=null, unstructuredReference=王乃成. 对GB/T 4458.6-2002的分析和讨论[J]. 机械工业标准化与质量, 2006(1): 32-35., articleTitle=对GB/T 4458.6-2002的分析和讨论, refAbstract=null), Reference(id=1263550871107977314, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, doi=null, pmid=null, pmcid=null, year=2006, volume=null, issue=1, pageStart=32, pageEnd=35, url=null, language=null, rfNumber=[20], rfOrder=28, authorNames=WANG N C, journalName=Mechanical Industry Standardization & Quality, refType=null, unstructuredReference=WANG N C. Analysis and discussion on GB/T 4458.6-2002[J]. Mechanical Industry Standardization & Quality, 2006(1): 32-35 (in Chinese)., articleTitle=Analysis and discussion on GB/T 4458.6-2002, refAbstract=null)], funds=[Fund(id=1263550836072956782, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, awardId=2023KTSCX327, language=EN, fundingSource=Characteristic Innovation Project of Ordinary Universities in Guangdong Province(2023KTSCX327), fundOrder=null, country=null), Fund(id=1263550837712929657, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, awardId=2023KTSCX327, language=CN, fundingSource=广东省普通高校特色创新类项目(2023KTSCX327), fundOrder=null, country=null), Fund(id=1263550838576956286, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, awardId=KY2023Y03Z, language=EN, fundingSource=Research Project of Zhuhai City Vocational and Technical College(KY2023Y03Z), fundOrder=null, country=null), Fund(id=1263550840275649420, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, awardId=KY2023Y03Z, language=CN, fundingSource=珠海城市职业技术学院科研项目(KY2023Y03Z), fundOrder=null, country=null), Fund(id=1263550841949176724, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, awardId=202304041101007, language=EN, fundingSource=Shanxi Province Science and Technology Cooperation and Exchange Special Project - Key National Science and Technology Cooperation Project(202304041101007), fundOrder=null, country=null), Fund(id=1263550844021162913, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, awardId=202304041101007, language=CN, fundingSource=山西省科技合作交流专项-国家重点科技合作项目(202304041101007), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1263550806947709520, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, xref=1, ext=[AuthorCompanyExt(id=1263550807098704465, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, companyId=1263550806947709520, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 College of Mechanical and Electrical Engineering, Zhuhai City Vocational and Technical College, Zhuhai Guangdong 519090, China), AuthorCompanyExt(id=1263550807186784850, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, companyId=1263550806947709520, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 珠海城市职业技术学院机电工程学院, 广东 珠海 519090)]), AuthorCompany(id=1263550807555883609, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, xref=2, ext=[AuthorCompanyExt(id=1263550807610409562, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, companyId=1263550807555883609, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 School of Automation and Software, Shanxi University, Taiyuan Shanxi 030006, China), AuthorCompanyExt(id=1263550807622992475, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, companyId=1263550807555883609, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 山西大学自动化与软件学院, 山西 太原 030006)])], figs=[ArticleFig(id=1263550820554031823, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, language=EN, label=Fig. 1, caption=Generative digital twin CAD-LDT, figureFileSmall=d/K0MK6pfeg9Oso8J4xJIQ==, figureFileBig=UxifmBWwoErqoqcNbuYoBA==, tableContent=null), ArticleFig(id=1263550820973462232, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, language=CN, label=图1, caption=生成式数字孪生体CAD-LDT, figureFileSmall=d/K0MK6pfeg9Oso8J4xJIQ==, figureFileBig=UxifmBWwoErqoqcNbuYoBA==, tableContent=null), ArticleFig(id=1263550821480973028, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, language=EN, label=Fig. 2, caption=Comparison of different methods for multimodal input scenarios ((a) Original image; (b) GPT-3.5+Adapter; (c) LLaVA-7B; (d) CAD-LDT), figureFileSmall=Yfm98cW9VsUCria5QDSGYw==, figureFileBig=G0Tfw97W2rgDnYZ0P2Rm2g==, tableContent=null), ArticleFig(id=1263550822605046514, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, language=CN, label=图2, caption=多模态输入场景不同方法的比较, figureFileSmall=Yfm98cW9VsUCria5QDSGYw==, figureFileBig=G0Tfw97W2rgDnYZ0P2Rm2g==, tableContent=null), ArticleFig(id=1263550824471511806, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, language=EN, label=Fig. 3, caption=Comparison of different methods in complex assembly scenarios ((a) Original image; (b) GPT-3.5+Adapter; (c) LLaVA-7B; (d) CAD-LDT), figureFileSmall=hTtSKbCocOekyQ4/BOCf3Q==, figureFileBig=foci083CkeRLpJ1YqIF16Q==, tableContent=null), ArticleFig(id=1263550824807056132, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, language=CN, label=图3, caption=复杂装配场景不同方法的比较, figureFileSmall=hTtSKbCocOekyQ4/BOCf3Q==, figureFileBig=foci083CkeRLpJ1YqIF16Q==, tableContent=null), ArticleFig(id=1263550825616556821, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, language=EN, label=Table 1, caption=

Data parameter table

, figureFileSmall=null, figureFileBig=null, tableContent=
参数项 配置值
优化器 AdamW
学习率 5e-5
批量大小 32
训练周期 50
正则化 0.01+0.1
), ArticleFig(id=1263550826497360677, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, language=CN, label=表1, caption=

数据参数表

, figureFileSmall=null, figureFileBig=null, tableContent=
参数项 配置值
优化器 AdamW
学习率 5e-5
批量大小 32
训练周期 50
正则化 0.01+0.1
), ArticleFig(id=1263550827361387308, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, language=EN, label=Table 2, caption=

Comparison table of model testing

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 IoU/% 约束满
足率/%
生成
时间/s
材料
利用率/%
最小壁厚
合格率/%
GPT-3.5+
Adapter
75.3 81.2 12.4 79.2 76.5
LLaVA-7B
CAD-LDT
79.1
83.6
85.7
91.3
9.8
7.2
86.7
92.4
82.3
94.1
), ArticleFig(id=1263550830150599488, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, language=CN, label=表2, caption=

孪生体测试对比表

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 IoU/% 约束满
足率/%
生成
时间/s
材料
利用率/%
最小壁厚
合格率/%
GPT-3.5+
Adapter
75.3 81.2 12.4 79.2 76.5
LLaVA-7B
CAD-LDT
79.1
83.6
85.7
91.3
9.8
7.2
86.7
92.4
82.3
94.1
), ArticleFig(id=1263550830679081801, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, language=EN, label=Table 3, caption=

Comparison table for complex assembly scenarios testing

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 特征完整性/
分数
关键尺寸
误差率/%
生成
时间/s
GPT-3.5+Adapter 75.3 18.8 32.5
LLaVA-7B 79.1 14.3 30.8
CAD-LDT 83.6 8.7 31.2
), ArticleFig(id=1263550831580857168, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, language=CN, label=表3, caption=

复杂装配场景测试对比表

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 特征完整性/
分数
关键尺寸
误差率/%
生成
时间/s
GPT-3.5+Adapter 75.3 18.8 32.5
LLaVA-7B 79.1 14.3 30.8
CAD-LDT 83.6 8.7 31.2
), ArticleFig(id=1263550833245995865, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, language=EN, label=Table 4, caption=

Comparison of twin experiments under different constraint conditions/%

, figureFileSmall=null, figureFileBig=null, tableContent=
消融项 IoU 约束满足率 失败率
完整模型 83.6 91.3 2.1
移除多模态数据
禁用约束编码器
70.9
76.2
82.5
71.4
15.7
27.3
传统空间表示 78.4 85.1 18.3
), ArticleFig(id=1263550833694786398, tenantId=1146029695717560320, journalId=1263187241531621409, articleId=1263514362317254680, language=CN, label=表4, caption=

不同约束条件下孪生体实验对比/%

, figureFileSmall=null, figureFileBig=null, tableContent=
消融项 IoU 约束满足率 失败率
完整模型 83.6 91.3 2.1
移除多模态数据
禁用约束编码器
70.9
76.2
82.5
71.4
15.7
27.3
传统空间表示 78.4 85.1 18.3
)], attaches=null, journal=Journal(id=1263185177418154013, delFlag=0, nameCn=图学学报, nameEn=Journal of Graphics, nameHistory1=null, nameHistory2=null, issn=2095-302X, eissn=null, cn=10-1034/T, coden=null, periodic=1, 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=o2S94GfCLjDYYJqsdQm5Mw==, journalPrice=null, startedYear=null, abbrevIsoEn=Journal of Graphics, journalRemark=null, publicationField=null, createdTime=1779096415144, updatedTime=1779097332356, createdBy=18614031015, updatedBy=13701087609, firstLetterCn=J, firstLetterEn=J, subjectCode=Natural Sciences, subjectName=null, subjectCodeEn=Natural Sciences, subjectNameEn=null, picCn=o2S94GfCLjDYYJqsdQm5Mw==, picEn=f7IBTa1ZXKPoIhMUwL1eVw==, jcr=null, cjcr=null, exts=[JournalExt(id=1263189024693469540, language=CN, name=图学学报, 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=1779097332404, updatedTime=1779097332404, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=http://www.txxb.com.cn/Journalx_txxb/authorLogOn.action, submissionEditorUrl=http://www.txxb.com.cn/Journalx_txxb/editorLogOn.action, submissionReviewUrl=http://www.txxb.com.cn/Journalx_txxb/expertLogOn.action, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1263189024764772709, language=EN, name=Journal of Graphics, 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=1779097332421, updatedTime=1779097332421, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=http://www.txxb.com.cn/Journalx_txxb/authorLogOn.action, submissionEditorUrl=http://www.txxb.com.cn/Journalx_txxb/editorLogOn.action, submissionReviewUrl=http://www.txxb.com.cn/Journalx_txxb/expertLogOn.action, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1263187241531621409, websiteList=[Website(id=1263188159811207473, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1263187241531621409, 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/txxb/CN, language=CN, createTime=1779097126200, createBy=18614031015, updateTime=1779097561850, updateBy=18614031015, name=图学学报-中文, tplId=1146099689490845704, title=图学学报, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1263194793929204110, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1263188159811207473, code=articleTextType, value=kx, createTime=1779098707897, updateTime=1779098707897, creator=18614031015, updator=18614031015), WebsiteProps(id=1263194793853706635, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1263188159811207473, code=banner, value=null, createTime=1779098707879, updateTime=1779098707879, creator=18614031015, updator=18614031015), WebsiteProps(id=1263194793966952849, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1263188159811207473, code=grayFlag, value=0, createTime=1779098707906, updateTime=1779098707906, creator=18614031015, updator=18614031015), WebsiteProps(id=1263194793824346506, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1263188159811207473, code=logo, value=https://castjournals.cast.org.cn/joweb/txxb/CN/file/pic?fileId=lmN7m2FoR6FhgnIimGeZBg==, createTime=1779098707872, updateTime=1779098707872, creator=18614031015, updator=18614031015), WebsiteProps(id=1263194793992118675, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1263188159811207473, code=minRunFlag, value=0, createTime=1779098707912, updateTime=1779098707912, creator=18614031015, updator=18614031015), WebsiteProps(id=1263194793891455373, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1263188159811207473, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/txxb/CN/file/pic, createTime=1779098707888, updateTime=1779098707888, creator=18614031015, updator=18614031015), WebsiteProps(id=1263194793979535762, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1263188159811207473, code=silenceFlag, value=0, createTime=1779098707909, updateTime=1779098707909, creator=18614031015, updator=18614031015), WebsiteProps(id=1263194793883066764, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1263188159811207473, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1779098707886, updateTime=1779098707886, creator=18614031015, updator=18614031015), WebsiteProps(id=1263194793933398415, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1263188159811207473, code=themeColor, value=null, createTime=1779098707898, updateTime=1779098707898, creator=18614031015, updator=18614031015), WebsiteProps(id=1263194793941787024, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1263188159811207473, code=themeStyle, value=null, createTime=1779098707900, updateTime=1779098707900, creator=18614031015, updator=18614031015)]), Website(id=1263188160184500536, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1263187241531621409, 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/txxb/EN, language=EN, createTime=1779097126289, createBy=18614031015, updateTime=1779097557941, updateBy=18614031015, name=图学学报-英文, tplId=1146101810881728533, title=Journal of Graphics, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1263194821443838360, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1263188160184500536, code=articleTextType, value=kx, createTime=1779098714457, updateTime=1779098714457, creator=18614031015, updator=18614031015), WebsiteProps(id=1263194821393506709, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1263188160184500536, code=banner, value=null, createTime=1779098714445, updateTime=1779098714445, creator=18614031015, updator=18614031015), WebsiteProps(id=1263194822144287131, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1263188160184500536, code=grayFlag, value=0, createTime=1779098714624, updateTime=1779098714624, creator=18614031015, updator=18614031015), WebsiteProps(id=1263194821334786452, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1263188160184500536, code=logo, value=https://castjournals.cast.org.cn/joweb/txxb/EN/file/pic?fileId=lmN7m2FoR6FhgnIimGeZBg==, createTime=1779098714431, updateTime=1779098714431, creator=18614031015, updator=18614031015), WebsiteProps(id=1263194822316253597, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1263188160184500536, code=minRunFlag, value=0, createTime=1779098714665, updateTime=1779098714665, creator=18614031015, updator=18614031015), WebsiteProps(id=1263194821414478231, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1263188160184500536, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/txxb/EN/file/pic, createTime=1779098714450, updateTime=1779098714450, creator=18614031015, updator=18614031015), WebsiteProps(id=1263194822249144732, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1263188160184500536, code=silenceFlag, value=0, createTime=1779098714649, updateTime=1779098714649, creator=18614031015, updator=18614031015), WebsiteProps(id=1263194821401895318, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1263188160184500536, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1779098714448, updateTime=1779098714448, creator=18614031015, updator=18614031015), WebsiteProps(id=1263194821716468121, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1263188160184500536, code=themeColor, value=null, createTime=1779098714522, updateTime=1779098714522, creator=18614031015, updator=18614031015), WebsiteProps(id=1263194821850685850, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1263188160184500536, code=themeStyle, value=null, createTime=1779098714554, updateTime=1779098714554, creator=18614031015, updator=18614031015)])], journalTitle=图学学报, weixinUrl=null, journalUrl=http://www.txxb.com.cn/, iacademicId=null, status=1, seqNo=null, journalTitleEn=Journal of Graphics, journalPhotoCn=o2S94GfCLjDYYJqsdQm5Mw==, journalPhotoEn=f7IBTa1ZXKPoIhMUwL1eVw==, journalFirstLetter=J, 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/txxb/CN/10.11996/JG.j.2095-302X.2026010173, detailUrlEn=https://castjournals.cast.org.cn/joweb/txxb/EN/10.11996/JG.j.2095-302X.2026010173, pdfUrlCn=https://castjournals.cast.org.cn/joweb/txxb/CN/PDF/10.11996/JG.j.2095-302X.2026010173, pdfUrlEn=https://castjournals.cast.org.cn/joweb/txxb/EN/PDF/10.11996/JG.j.2095-302X.2026010173, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于大模型的生成式数字孪生体建模
收藏切换
PDF下载
梁生龙 1 , 范秋霞 2
图学学报 | 数字化设计与制造 2026,47(1): 173-178
收起
收藏切换
图学学报 | 数字化设计与制造 2026, 47(1): 173-178
基于大模型的生成式数字孪生体建模
全屏
梁生龙1 , 范秋霞2
作者信息
  • 1 珠海城市职业技术学院机电工程学院, 广东 珠海 519090
  • 2 山西大学自动化与软件学院, 山西 太原 030006

通讯作者:

梁生龙,E-mail:
Generative digital twin modeling based on large models
Shenglong LIANG1 , Qiuxia FAN2
Affiliations
  • 1 College of Mechanical and Electrical Engineering, Zhuhai City Vocational and Technical College, Zhuhai Guangdong 519090, China
  • 2 School of Automation and Software, Shanxi University, Taiyuan Shanxi 030006, China
出版时间: 2026-02-28 doi: 10.11996/JG.j.2095-302X.2026010173
文章导航
收藏切换

针对数字孪生(DT)技术与大模型在工业设计中融合应用的挑战,提出了一种基于生成式大模型的CAD-LDT数字孪生体建模框架。通过构建物理实体模块、智能生成模块和虚拟实体模块的三元架构,创新性地整合多模态数据融合机制与领域知识约束,实现从物理实体描述到参数化CAD模型的自主生成。采用LLaVA-7B和LLaMA-7B作为基础模型,通过LoRA轻量化适配器实现视觉-文本特征的跨模态对齐,并设计约束编码器将几何公差与物理规则转化为结构化JSON对象。为增强空间变换的数学一致性,引入李群算法优化刚体变换表征;采用几何权重分箱方法离散化复杂装配关系;提出时空解耦生成策略,协同规划空间布局与装配时序。在DeepCAD数据集上的实验结果表明,该框架在几何达到83.6%,约束满足率达91.3%,生成效率提升了26.5%,显著优于主流基线模型。消融实验进一步验证了多模态融合、约束编码机制和李群算法对建模质量的关键贡献。为智能制造领域提供了新的DT建模范式,在参数化设计、装配工艺优化等方面展现出工程应用价值。

大模型  /  数字孪生  /  多模态数据  /  智能制造  /  参数化设计

To address the challenges in integrating Digital-Twin (DT) technology with large-scale generative models in industrial design, a CAD-LDT digital-twin modeling framework based on generative foundation models was proposed. The framework adopted a triadic architecture consisting of a physical-entity module, an intelligent generation module, and a virtual-entity module, and innovatively incorporated multi-modal data fusion mechanisms and domain-knowledge constraints to enable autonomous generation of parameterized CAD models from physical-entity descriptions. Utilizing LLaVA-7B and LLaMA-7B as backbone models, the framework employed LoRA-based lightweight adapters to achieve cross-modal alignment between visual and textual features, and introduced a constraint encoder that transformed geometric tolerances and physical rules into structured JSON objects. To enhance the mathematical consistency of spatial transformations, Lie-group algorithms were adopted for the optimization of rigid-body transformations, while a geometric-weight binning strategy was proposed to discretize complex assembly relationships. Moreover, a spatiotemporal-decoupled generation strategy was designed to jointly optimize spatial layout and assembly sequencing. Experimental results on the DeepCAD dataset indicated that the proposed framework achieved an Intersection- over-Union (IoU) of 83.6%, a constraint satisfaction rate of 91.3%, and a 26.5% improvement in generation efficiency, significantly outperforming existing baseline models. Further ablation studies confirmed the critical contributions of multi-modal fusion, constraint encoding mechanisms, and Lie-group optimization to modeling performance, providing a novel DT modeling paradigm for intelligent manufacturing with demonstrated value in parametric design and assembly process optimization.

large models  /  digital twin  /  multimodal data  /  intelligent manufacturing  /  parametric design
梁生龙, 范秋霞. 基于大模型的生成式数字孪生体建模. 图学学报, 2026 , 47 (1) : 173 -178 . DOI: 10.11996/JG.j.2095-302X.2026010173
Shenglong LIANG, Qiuxia FAN. Generative digital twin modeling based on large models[J]. Journal of Graphics, 2026 , 47 (1) : 173 -178 . DOI: 10.11996/JG.j.2095-302X.2026010173
随着计算机软硬件的高速发展,数字孪生(Digital Twin, DT)技术在工业领域也得到了越来越多的应用。DT是物理产品的虚拟和数字化复制品,是一组信息集,能够在细节和整体上全面反映物理产品的特性[1-2];王进峰等[3]认为DT是指将实体工厂完美地以一比一的比例复制到虚拟环境中,通过实时收集和分析数据,并同步至虚拟工厂中进行处理,进而对实际工厂产生影响,这一概念主要聚焦于增强工业现场的任务执行效率、使生产装配流程更加透明化,以及提高作业安全性;文献[4-6]认为DT仿真需要对应物理实体运行中的状态和特性,然而其侧重于状态监测而非主动生成,缺乏对设计需求的逆向建模能力;TAO等[7]运用DT在CPS物理空间和信息空间进行数据的不断交互,物理空间根据信息空间模型的仿真和优化结果进行控制和调整,推动数控装备的可靠性分析。
大模型的概念近年来在自然语言处理领域引起了广泛关注,并在众多应用中展示了其强大的表现能力。Transformer模型通过自注意力机制(Self-attention mechanism)有效地捕捉文本中远程依赖关系,使得模型能够更好地理解上下文信息[8]。这一特性使得LLM(Large Language Model)能够在缺乏大规模标注数据的情况下,仍能表现出卓越的性能。此外,LLM还可以通过少样本学习和零样本学习的方式来适应新的任务,其极大地拓宽了其应用场景[9]。这种架构不仅在NLP (Natural Language Processing)领域得到了广泛应用,还逐渐被引入到其他领域,如计算机视觉和语音识别等[10]。YAO等[11]提出了用于文本游戏中动作生成的上下文动作语言模型。PANG等[12]将预训练的语言模型与变分自编码器相结合,用于大模型指令翻译,把自然语言指令转换为任务语言,但现有LLM在工业场景的应用多局限于文本指令翻译,尚未实现多模态数据与领域知识的深度融合。
DT和大模型技术在多个领域的应用取得了大量研究成果,并展示了强大的能力,但DT结合LLM技术在工业设计中的实际应用仍面临挑战。
本文介绍了一种CAD-LDT (Computer-Aided Design with Large-Model Based Digital Twin)数字孪生体,从CAD设计机理入手,结合DT技术及大模型根据物理实体和要求实现自主生成所需的模型。通过利用多模态数据训练专用于CAD设计领域的大模型,从物理实体中提取特定的知识,以及通过人工所提出的设计需求作为语料与大模型进行孪生模型的生成,已实现高效的CAD建模。
基本的模型数字孪生体如图1所示,主要分为3个模块,①带有几何约束[13]与物理约束[14]的物理实体模块;②微调后的大模型以及多模态数据组成的智能生成模块;③相应CAD零件组成的装配体的虚拟实体模块。物理实体根据其特性可以分为几何约束与物理约束,二者共同构建了虚实映射的工程基准,通过采取多模态数据融合策略,利用激光扫描仪和工业相机传感器获取物理模型的特征,采集高精度三维几何特征和多视角2D图像来提取轮廓特征。其具体数据形式包括了用于重建物理实体的边界表示模型的点云数据、2000×1500分辨率的多视角图像数据和相关工艺文档,包含材料属性和装配公差等物理约束。几何约束表示物理实体通过点/线/面拓扑关系定义形状边界,如齿轮的渐开线齿廓、轴孔同心度等,并采取参数化表达的方式,如直径D=200 mm ± 0.1 mm,为后续CAD建模提供精确数学描述;物理约束实体是在几何约束基础上叠加材料属性,形成可验证的物理规则,确保生成模型满足强度等要求。
智能生成模块中多模态数据融合实现约束解算,多模态数据管道图像/视频流经LoRA(Low-Rank Adaptation)[15]轻量化适配器,提取轮廓特征,而文本指令则通过LLM解析为结构化操作符。同时提供了一个约束编码器,将几何约束与物理约束编码为JSON对象,部分字段定义样例如下:
{
“几何约束”:
{
“类型”:“同心度”,
“基准元素”:“轴A”,
“目标元素”:“孔B”,
“公差”:“0.05 mm”
},
“物理约束”:
{
“类型”:“材料强度”,
“材料牌号”:“45钢”,
“屈服强度”:“355 MPa”,
“安全系数”:1.5
}
}
最终输出参数化CAD模型及其装配关系,具体包括几何参数和装配关系这些结构化参数序列,以及公差调整和装配序列规划的设计优化建议。虚拟实体模块内部图像引擎采用CadQuery[16]内核引擎,能够基于JSON约束动态生成参数化CAD模型,以及支持STEP/IGES等标准格式输出。经过物理实体模块和智能生成模块所生成的CAD组件进行一定装配后构成最终的装配体。
该模型架构采用了LLaVA-7B和预训练后的LLaMA-7B作为基础模型。在微调部分,对于输入图像,利用预先训练的视觉编码器ViT(Vision Transformer)提供视觉功能。并将一个简单的2层线性层视为视觉-语言投影器,将视觉补丁嵌入映射到文本特征空间。因此,模型中有一个视觉标记序列,可以像LLM的文本标记一样进行理解。具体来说,对于长度为K的序列,可通过下式计算目标答案的概率,即
$p({S}_{a})={\displaystyle \prod _{k=1}^{K}{p}_{\text{local}}({S}_{k})\cdot {p}_{\text{global}}\left(Assembly\left|\right\{{S}_{k}\}\right)}$
式中:局部项为
${p}_{\text{local}}({x}_{i})\propto \mathrm{exp}\left(W\cdot {h}_{i}+\lambda {E}_{\text{global}}({x}_{i})\right)$
式中:Eglobal表示当前token对全局装配可行性的预评估。
全局项则基于超图注意力[17]建模多零件关系,其定义为
${p}_{\text{global}}=\frac{1}{\text{Z}}\mathrm{exp}\left(-{\displaystyle \sum _{c\in C}{\Phi }_{c}(违反约束C)}\right)$
式中:Z表示归一化常数;Φc表示约束违反惩罚函数。为避免局部生成与全局约束冲突,显式建模零件间依赖关系。同时,装配体需处理空间布局和时间顺序,解决传统单序列生成难以解耦的问题,并将生成序列分成了空间序列和时间序列2部分,其联合概率为
$p\left({S}_{a}\right)=p\left({S}_{\text{space}}\right)\cdot p\left({S}_{\text{time}}|{S}_{\text{space}}\right)$
式中:Sspace表示空间布局序列;Stime表示时间布局序列。
在子模型设计方面,在空间布局上可使用扩散模型[18]生成初始零件位置分布,其具体建模方法为
$P({S}_{\text{space}})={\displaystyle \prod _{i=1}^{{L}_{s}}p\left({x}_{i}^{\text{space}}|全局包围盒,\text{ }\text{ }{S}_{\text{instruct}}\right)}$
同时可在时间顺序方面引入了物理引擎模拟,以评估安装工具的可达性,即
$p({S}_{\text{time}}|{S}_{\text{space}})={\displaystyle \prod _{j=1}^{{L}_{t}}p\left({x}_{j}^{\text{time}}|{S}_{\text{space}},\text{ }\text{ }物理可达性\right)}$
后续在空间模型预训练中基于无装配顺序的零件位置数据,以及对时间模型进行微调,完成分离空间规划与动作序列。
传统方法如VAE(Variational Autoe-Ecoder)和VQ-VAE(Vector Quantized Variational Auto Encoder)等依赖潜在向量或点云作为输入,需专业设备进行采集,其数据获取过程复杂且昂贵,且潜在向量的抽象性导致生成结果依赖输入质量,易产生累积误差。因此本文在框架中加入了李群算法[19]等用来增强大模型的空间推理能力,又由于传统均匀分箱方法无法适应复杂装配体的几何特征,特使用了几何权重分箱方法进行空间关系离散化,并设零件表面点云为P={pi},对每个离散方向θ,定义分箱权重为
$\omega (\theta )=\frac{1}{N}{\displaystyle \sum _{{p}_{i}\in \rho }\kappa \left({p}_{i}\right)\cdot \text{δ}\left(\theta -\theta \left({p}_{i}\right)\right)}$
式中:κ(pi)表示点处的曲率,用于增强高曲率区域的离散密度;δ(·)表示Dirac函数,筛选属于当前分箱方向的点。
在装配约束编码部分使用李群嵌入来解决传统欧拉角/平移分离表示难以保证刚体变换的数学一致性,并设零件A到零件B的相对位姿为TAB$\in$SE(3),并将其映射到切空间,即
$\xi_{A B}=\log \left(T_{A B}\right) \in \text { æ }(3)$
然后对切空间向量${\xi }_{AB}=[\omega,\text{ }\text{ }\upsilon ]T$进行分层离散化,旋转分量ω$\in$R3映射到四元数球面网格。平移分量υ$\in$R3按零件包围盒尺寸自适应量化。
本文选择的是DeepCAD数据集,该数据集包含约17.8万单个零件的建模序列。首先,使用CADQuery图形库,从固定角度为每个 CAD模型渲染2D图像。然后,开发了需要给模型输入几种不同的自然语言建模指令,旨在指导CAD-LDT根据参考图像生成CAD模型。在每次微调迭代中,随机选择一个不同的指令。使用预训练的视觉编码器提取视频和图片中的图像特征,与离散化token拼接为多模态输入序列。而对描述性文本进行语义解析,生成结构化参数序列。最后将数据集分成了训练-验证-测试集,按8∶1∶1划分,测试集覆盖了极端案例(如遮挡图像、模糊文本)以验证鲁棒性。
然后先对原始CAD模型的B-Rep(Boundary Representation)边界JSON格式表示数据和渲染图像进行联合推理,剔除无法生成有效描述如非流形曲面等的非常规几何结构。随后基于InstructGPT构建参数化指令生成管道,结合GB/T 4458[20]机械制图标准术语生成包含尺寸公差和功能属性的结构化描述文本。通过三阶段人工校验流程:①删除与STEP(Standard for the Exchange of Product Model Data)文件中几何矛盾的描述;②轮修正不符合标准的尺寸标注;③保留18 000条精准描述并量化错误率低于0.3% (95%置信区间)。生成的自然语言描述与CADQuery渲染图像、参数化代码构成三元组数据,按LLaVA兼容格式存储用于跨模态对比学习,其中文本指令通过LLM解析器转换为结构化操作符,与视觉编码器提取的图像特征共同输入多模态融合模块。
为了全面评估预测序列,本文采用了一组指标来评估预测的不同方面。具体而言,最终的CAD重建通过使用适用几何匹配度评估的交并比(Intersection over Union, IoU)与约束满足率进行定量分析。由于CAD序列是作为标记进行预测的,当使用CADQuery进行重建时,其并不总是生成成功渲染的CAD模型,因此本文引入了一个约束满足率指标,以百分比表示,用于量化无效模型的比例,反映生成方案对几何约束、工艺约束和标注规范的满足程度。此外,还使用生成时间指标评估实时性,由于生成时间直接影响设计迭代周期,在实际的应用中可缩短产品开发周期。
表1可见,在实验过程中,优化器为AdamW,为了避免陷入尖锐局部极小值,需要将学习率设置为5e-5,而批量大小设置为32,使其能够兼顾显存效率与梯度稳定性,训练周期为50个epochs。同时也需要进行正则化,此时权重衰减值为0.01防止过拟合,而0.1的Dropout值则为了提升泛化性。
为了验证所提框架的有效性,本文在DeepCAD测试集上对GPT-3.5+Adapter,LLaVA-7B和CAD- LDT的3种方法进行了对比实验。表2实验结果表明,CAD-LDT在IoU、约束满足率和生成时间3个核心指标上均显著优于基线模型。具体而言,在几何匹配度方面,CAD-LDT的IoU为83.6%较LLaVA-7B提升4.5%,说明多模态数据融合策略有效增强了模型对几何拓扑关系的理解能力。通过图2可视化分析发现,本方法在渐开线齿廓、同心轴孔等复杂几何特征的建模精度上具有显著优势。
而91.3%的约束满足率验证了约束编码器与李群算法的协同作用,相比其他2种模型的81.2%和85.7%的满足率有了明显地提高。更为重要的是,在设计质量指标上,92.4%的材料利用率较LLaVA-7B提升了5.7个百分点,表明模型通过轮廓优化排样有效减少材料浪费,其最小壁厚合格率也远超基线的82.3%,证明物理规则编码成功规避薄壁缺陷。在生成时间方面,7.2 s的平均生成时间较LLaVA-7B缩短26.5%,这得益于参数化CAD内核的轻量化部署,LoRA轻量化适配器将视觉特征提取耗时降低58.0%,同时约束编码器的JSON结构化输出使CADQuery建模效率提升32.0%。
为验证复杂系统装配能力,利用车铣复合数控机床的相关核心子系统作为测试案例(含横梁、导轨、主轴箱等零件),由表3可看出,GPT-3.5+ Adapter模型的特征完整性评分为75.3分,关键尺寸误差率为18.8%,生成时间为32.5 s。LLaVA-7B模型在特征完整性方面有所提升,关键尺寸误差率降低至14.3%,生成时间缩短至30.8 s。值得注意的是,CAD-LDT模型展现出显著优势:特征完整性达到83.6分的最高水平,关键尺寸误差率控制在8.7%的最优值,生成时间维持在31.2 s的可接受范围,其可视化图如图3所示。
约束编码器注入的结构化规则使关键尺寸误差率降低10.1个百分点,而LoRA轻量化适配器通过剪枝冗余视觉特征提取路径,在保持特征完整性的同时有效控制计算耗时。尽管CAD-LDT生成时间略高于LLaVA-7B,但其在特征完整性与尺寸精度上的显著提升,证明该折衷策略对高精度装配场景具有决定性价值。
在消融实验部分,本文针对模型的3种关键组件进行消融实验,第1组移除了多模态数据,第2组禁用框架内部的约束编码器,第3组采用传统欧拉角替代李群算法,进行传统空间表示。
实验结果见表4。在移除多模态数据后,IoU相比原来完整的框架下降了12.7%,表明图像特征与文本指令的跨模态对齐对几何理解具有决定性作用。而禁用了约束编码器后,会导致所生成的模型约束满足率骤降至71.4%,说明结构化约束表示对工程可行性具有比较高的重要性。在最后一组中,采用传统欧拉角替代李群算法时,模型失败率上升至18.3%,足以证明刚体变换的数学一致性对复杂装配至关重要。而在具有完整模块的情况下,其失败率仍有2.1%,在所有失败案例中68%源于点云特征缺失(如高反光表面扫描空洞),32%因过盈配合与热膨胀约束矛盾等可导致公差冲突。在后续的研究中,将尝试引入多传感器融合补偿点云,增加约束优先级仲裁模块。
本文提出了一种基于大模型的生成式数字孪生体建模框架CAD-LDT,通过融合多模态数据与领域知识约束,实现了从物理实体描述到参数化CAD模型的自主生成。实验表明,该框架在几何匹配度、约束满足率及生成效率等核心指标上显著优于现有基线模型。在该框架中,多模态融合机制(LoRA适配器+约束编码器)使几何交并比提升至83.6%,在DeepCAD数据集上91.3%的约束满足率验证了结构化知识表达的工程可行性;李群算法与几何权重分箱显著优化复杂装配关系表征,将空间变换一致性错误率降至2.1%;时空解耦生成策略实现空间布局与装配时序协同规划,生成效率达7.2秒/件(提速26.5%),在某数控企业实际应用显示工艺变更响应时间缩短40.0%。上述结果证明:多模态融合机制有效解决视觉-文本跨模态对齐问题,李群算法保障了数学表征的物理一致性,时空解耦策略突破传统序列模型的耦合局限。
然而,本研究仍存在若干待改进之处:首先,当前训练数据主要聚焦机械零件级建模,仅在数控机床这一对象上进行了测试,对整机系统级装配场景的覆盖不足;其次,模型对极端工艺约束的泛化能力有待验证。未来研究将从以下方向展开:
1) 扩展工业级多尺度装配数据集,开发层次化生成框架以支持从零件到系统的全流程建模;
2) 结合物理信息神经网络强化模型对复杂力学约束的嵌入能力;
3) 构建数字孪生体在线更新机制,通过边缘计算设备实现物理-虚拟空间的毫秒级双向反馈;
4) 探索大模型与增强现实技术的融合应用,开发支持自然语言交互的智能化设计辅助系统。随着多模态大模型与数字线程技术的深度融合,生成式DT有望在个性化定制、故障预测性维护等领域产生更大的应用价值。
  • 广东省普通高校特色创新类项目(2023KTSCX327)
  • 珠海城市职业技术学院科研项目(KY2023Y03Z)
  • 山西省科技合作交流专项-国家重点科技合作项目(202304041101007)
参考文献 引证文献
排序方式:
[1]
王剑, 王好臣, 李学伟, . 基于OPC UA的数字孪生车间信息物理融合系统[J]. 现代制造工程, 2023(4): 43-50.
WANG J, WANG H C, LI X W, et al. Digital twin workshop information physical fusion system based on OPC UA[J]. Modern Manufacturing Engineering, 2023(4): 43-50 (in Chinese).
[2]
GRIEVES M, VICKERS J. Digital twin: mitigating unpredictable, undesirable emergent behavior in complex systems[M]// KAHLEN F J, FLUMERFELT S, ALVES A. Transdisciplinary Perspectives on Complex Systems:New Findings and Approaches. Cham: Springer, 2017: 85-113.
[3]
王进峰, 问丛川, 花广如. 面向概念、技术与应用的数字孪生综述[J]. 中国工程机械学报, 2023, 21(2): 112-116, 133.
WANG J F, WEN C C, HUA G R. A survey of digital twins for concept, technology and application[J]. Chinese Journal of Construction Machinery, 2023, 21(2): 112-116, 133 (in Chinese).
[4]
LEE J, LAPIRA E, YANG S H, et al. Predictive manufacturing system-trends of next-generation production systems[J]. IFAC Proceedings Volumes, 2013, 46(7): 150-156.
[5]
BAO J S, GUO D S, LI J, et al. The modelling and operations for the digital twin in the context of manufacturing[J]. Enterprise Information Systems, 2019, 13(4): 534-556.
[6]
ZHUANG C B, LIU J H, XIONG H. Digital twin-based smart production management and control framework for the complex product assembly shop-floor[J]. The International Journal of Advanced Manufacturing Technology, 2018, 96(1/4): 1149-1163.
[7]
TAO F, ZHANG M, LIU Y S, et al. Digital twin driven prognostics and health management for complex equipment[J]. CIRP Annals, 2018, 67(1): 169-172.
[8]
JIANG Y K, CHEN J H, ZHOU H C, et al. Contour error modeling and compensation of CNC machining based on deep learning and reinforcement learning[J]. The International Journal of Advanced Manufacturing Technology, 2022, 118(1/2): 551-570.
[9]
SAMSONOV V, CHRISMARIE E, KÖPKEN H G, et al. Deep representation learning and reinforcement learning for workpiece setup optimization in CNC milling[J]. Production Engineering, 2023, 17(6): 847-859.
[10]
LI B R, ZHANG H, YE P Q, et al. Trajectory smoothing method using reinforcement learning for computer numerical control machine tools[J]. Robotics and Computer-Integrated Manufacturing, 2020, 61: 101847.
[11]
YAO S Y, RAO R H, HAUSKNECHT M, et al. Keep CALM and explore: language models for action generation in text-based games[C]// 2020 Conference on Empirical Methods in Natural Language Processing. Stroudsburg: ACL, 2020: 8736-8754.
[12]
PANG J C, YANG X Y, YANG S H, et al. Natural language-conditioned reinforcement learning with inside-out task language development and translation[EB/OL]. [2024- 04-10]. https://arxiv.org/pdf/2302.09368. https://arxiv.org/pdf/2302.09368
[13]
何援军, 于海燕, 方志刚. 基于形计算重构CAD的计算基础[J]. 计算机集成制造系统, 2025, 31(3): 760-777.
HE Y J, YU H Y, FANG Z G. Reconstructing computational foundation for CAD based on shape computing[J]. Computer Integrated Manufacturing Systems, 2025, 31(3): 760-777 (in Chinese).
[14]
黄川, 李雅琴, 祁越然, . 基于3D-CAE的高光谱解混及小样本分类方法[J]. 自然资源遥感, 2025, 37(1): 8-14.
HUANG C, LI Y Q, QI Y R, et al. A hyperspectral unmixing and few-shot classification method based on 3DCAE network[J]. Remote Sensing for Natural Resources, 2025, 37(1): 8-14 (in Chinese).
[15]
龙禹辰, 勾智楠, 陈宇欣, . 基于大语言模型的多任务生成式重构对话情绪识别[J]. 计算机应用研究, 2025, 42(7): 1964-1971.
LONG Y C, GOU Z N, CHEN Y X, et al. Multi-task generative emotion recognition in conversation based on large language models[J]. Application Research of Computers, 2025, 42(7): 1964-1971 (in Chinese).
[16]
URBANCZYK A, WRIGHT J. CadQuery: a python parametric CAD scripting framework based on OCCT[EB/OL]. [2025- 04-10]. https://github.com/CadQuery/cadquery. https://github.com/CadQuery/cadquery
[17]
刘静文, 刘渊, 袁琮淇. 融合超图卷积和自监督协同训练的组推荐算法[J]. 中文信息学报, 2024, 38(7): 115-126, 136.
LIU J W, LIU Y, YUAN C Q. Group recommendation algorithms incorporating hypergraph convolution and self-supervised collaborative training[J]. Journal of Chinese Information Processing, 2024, 38(7): 115-126, 136 (in Chinese).
[18]
雷松林, 赵征鹏, 阳秋霞, . 基于可解耦扩散模型的零样本风格迁移[EB/OL]. (2025-03-31) [2025-04-02]. http://kns.cnki.net/kcms/detail/10.1034.T.20250331.1207.002.html. http://kns.cnki.net/kcms/detail/10.1034.T.20250331.1207.002.html
LEI S L, ZHAO Z P, YANG Q X, et al. Zero-shot style transfer based on decoupled diffusion models[EB/OL]. (2025-03-31) [2025-04-02]. http://kns.cnki.net/kcms/detail/10.1034.T.20250331.1207.002.html in Chinese). http://kns.cnki.net/kcms/detail/10.1034.T.20250331.1207.002.html
[19]
孙灏铖, 刘力, 李凡长. 李群模糊C均值聚类图像分割算法[J]. 软件学报, 2024, 35(10): 4806-4825.
SUN H C, LIU L, LI F Z. Lie group fuzzy C-means clustering algorithm for image segmentation[J]. Journal of Software, 2024, 35(10): 4806-4825 (in Chinese).
[20]
王乃成. 对GB/T 4458.6-2002的分析和讨论[J]. 机械工业标准化与质量, 2006(1): 32-35.
WANG N C. Analysis and discussion on GB/T 4458.6-2002[J]. Mechanical Industry Standardization & Quality, 2006(1): 32-35 (in Chinese).
2026年第47卷第1期
PDF下载
15
5
引用本文
BibTeX
文章信息
doi: 10.11996/JG.j.2095-302X.2026010173
  • 接收时间:2025-04-13
  • 首发时间:2026-05-19
  • 出版时间:2026-02-28
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2025-04-13
  • 录用日期:2025-08-08
基金
Characteristic Innovation Project of Ordinary Universities in Guangdong Province(2023KTSCX327)
广东省普通高校特色创新类项目(2023KTSCX327)
Research Project of Zhuhai City Vocational and Technical College(KY2023Y03Z)
珠海城市职业技术学院科研项目(KY2023Y03Z)
Shanxi Province Science and Technology Cooperation and Exchange Special Project - Key National Science and Technology Cooperation Project(202304041101007)
山西省科技合作交流专项-国家重点科技合作项目(202304041101007)
作者信息
    1 珠海城市职业技术学院机电工程学院, 广东 珠海 519090
    2 山西大学自动化与软件学院, 山西 太原 030006

通讯作者:

梁生龙,E-mail:
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/txxb/CN/10.11996/JG.j.2095-302X.2026010173
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