Article(id=1217789891908391845, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1217789884081820362, articleNumber=null, orderNo=null, doi=10.12404/j.issn.1671-1815.2407675, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=research-article, receivedDate=1728921600000, receivedDateStr=2024-10-15, revisedDate=1745251200000, revisedDateStr=2025-04-22, acceptedDate=null, acceptedDateStr=null, onlineDate=1768273335672, onlineDateStr=2026-01-13, pubDate=1753632000000, pubDateStr=2025-07-28, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1768273335672, onlineIssueDateStr=2026-01-13, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1768273335672, creator=13701087609, updateTime=1768273335672, updator=13701087609, issue=Issue{id=1217789884081820362, tenantId=1146029695717560320, journalId=1146123166801305609, year='2025', volume='25', issue='21', pageStart='8761', pageEnd='9209', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=null, createTime=1768273333807, creator=13701087609, updateTime=1768273602927, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1217791012932604619, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1217789884081820362, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1217791012932604620, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1217789884081820362, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=8945, endPage=8954, ext={EN=ArticleExt(id=1217789893330260968, articleId=1217789891908391845, tenantId=1146029695717560320, journalId=1146123166801305609, language=EN, title=Data-driven Transient Stability Assessment of Power Systems Based on Transformer-GAT Parallel Feature Fusion, columnId=1156262733675876713, journalTitle=Science Technology and Engineering, columnName=Papers·Electrical Technology, runingTitle=null, highlight=null, articleAbstract=

With the rapid development of the power system, the large-scale integration of new energy into the grid and the coordinated optimization of source-grid-load-storage have increased the proportion of power electronic equipment, making the stability of the power grid, especially the assessment of transient stability, particularly important. Aiming at the problem of insufficient consideration of topological structure in traditional methods, a deep learning method based on Transformer-graph attention network(GAT) parallel feature fusion was proposed for the transient stability evaluation of power systems. The busbar voltage amplitude, phase angle and topological adjacentation matrix were taken as input features. Batch data were generated using the Siemens simulation software PSS/E, and features were extracted in parallel through Transformer and GAT. Weighted fusion was carried out using the attention mechanism. The comparison results with other methods show that this method simulates different load conditions and fault conditions in the IEEE 39-node system. The results indicate that the evaluation accuracy and robustness are relatively high, and it can effectively improve the safety and stability of the power system.

, correspAuthors=Tian-qi XU, 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=Yi-fan HOU, Tian-qi XU, Yan LI, Xiao-lan LI), CN=ArticleExt(id=1217789897180631334, articleId=1217789891908391845, tenantId=1146029695717560320, journalId=1146123166801305609, language=CN, title=基于Transformer-GAT并行特征融合的数据驱动的电力系统暂态稳定评估, columnId=1156262734506353627, journalTitle=科学技术与工程, columnName=论文·电工技术, runingTitle=null, highlight=null, articleAbstract=

随着电力系统的快速发展,大规模新能源并网及源网荷储协同优化增加了电力电子设备的比重,使得电网稳定性,特别是暂态稳定性评估,变得尤为重要。针对传统方法在拓扑结构考虑不足的问题,提出了一种基于Transformer-图注意力网络(graph attention network,GAT)并行特征融合的深度学习方法,用于电力系统暂态稳定性评估。以母线电压幅值、相角及拓扑邻接矩阵作为输入特征,利用西门子仿真软件PSS/E生成批量数据,并通过Transformer和GAT并行提取特征,采用注意力机制进行加权融合。与其他方法的对比结果表明,该方法在IEEE 39节点系统中模拟了不同负荷条件和故障工况,结果表明评估精度和鲁棒性较高,能够有效提升电力系统的安全稳定性。

, correspAuthors=徐天奇, authorNote=null, correspAuthorsNote=
* 徐天奇(1978—),男,汉族,云南禄丰人,博士,教授。研究方向:韧性电网、新能源发电并网、电力信息物理系统。E-mail:
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=oeV88547YMLIu3Hx65D9FA==, magXml=Ej2uaDplq/ldy96LWlgF8A==, pdfUrl=null, pdf=As5nJ6vk8OvAwgMyuas8RA==, pdfFileSize=6240381, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=Zpcpa0XotWwa91jY6VkQMQ==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=4m4JWczz/gdiuy1w0WhNEA==, mapNumber=null, authorCompany=null, fund=null, authors=

侯一帆(2000—),男,汉族,甘肃庆阳人,硕士研究生。研究方向:数据驱动电力系统。E-mail:

, authorsList=侯一帆, 徐天奇, 李琰, 李晓兰)}, authors=[Author(id=1217860109326270626, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=2856334034@qq.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1217860109456294065, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, authorId=1217860109326270626, language=EN, stringName=Yi-fan HOU, firstName=Yi-fan, middleName=null, lastName=HOU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 Key Laboratory of Cyber-Physical Power System of Yunnan Universities, School of Electrical and Information Technology, Yunnan Minzu University, Kunming 650504, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1217860109577928898, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, authorId=1217860109326270626, language=CN, stringName=侯一帆, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 云南民族大学电气信息工程学院云南省高校电力信息物理融合系统重点实验室, 昆明 650504, bio={"content":"

侯一帆(2000—),男,汉族,甘肃庆阳人,硕士研究生。研究方向:数据驱动电力系统。E-mail:

"}, bioImg=null, bioContent=

侯一帆(2000—),男,汉族,甘肃庆阳人,硕士研究生。研究方向:数据驱动电力系统。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1217860108982337665, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, xref=1, ext=[AuthorCompanyExt(id=1217860108999114882, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, companyId=1217860108982337665, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Key Laboratory of Cyber-Physical Power System of Yunnan Universities, School of Electrical and Information Technology, Yunnan Minzu University, Kunming 650504, China), AuthorCompanyExt(id=1217860109015892100, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, companyId=1217860108982337665, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 云南民族大学电气信息工程学院云南省高校电力信息物理融合系统重点实验室, 昆明 650504)])]), Author(id=1217860109779255507, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=xu.tianqi@ymu.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1217860109993165037, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, authorId=1217860109779255507, language=EN, stringName=Tian-qi XU, firstName=Tian-qi, middleName=null, lastName=XU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, *, address=1 Key Laboratory of Cyber-Physical Power System of Yunnan Universities, School of Electrical and Information Technology, Yunnan Minzu University, Kunming 650504, China
2 Yunnan Key Laboratory of Unmanned Autonomous System, Kunming 650504, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1217860110089634039, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, authorId=1217860109779255507, language=CN, stringName=徐天奇, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, *, address=1 云南民族大学电气信息工程学院云南省高校电力信息物理融合系统重点实验室, 昆明 650504
2 云南省无人自主系统重点实验室, 昆明 650504, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1217860108982337665, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, xref=1, ext=[AuthorCompanyExt(id=1217860108999114882, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, companyId=1217860108982337665, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Key Laboratory of Cyber-Physical Power System of Yunnan Universities, School of Electrical and Information Technology, Yunnan Minzu University, Kunming 650504, China), AuthorCompanyExt(id=1217860109015892100, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, companyId=1217860108982337665, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 云南民族大学电气信息工程学院云南省高校电力信息物理融合系统重点实验室, 昆明 650504)]), AuthorCompany(id=1217860109124944015, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, xref=2, ext=[AuthorCompanyExt(id=1217860109145915539, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, companyId=1217860109124944015, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Yunnan Key Laboratory of Unmanned Autonomous System, Kunming 650504, China), AuthorCompanyExt(id=1217860109154304148, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, companyId=1217860109124944015, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 云南省无人自主系统重点实验室, 昆明 650504)])]), Author(id=1217860111398256899, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, 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=1217860111557640463, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, authorId=1217860111398256899, language=EN, stringName=Yan LI, firstName=Yan, middleName=null, lastName=LI, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 Key Laboratory of Cyber-Physical Power System of Yunnan Universities, School of Electrical and Information Technology, Yunnan Minzu University, Kunming 650504, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1217860111696052510, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, authorId=1217860111398256899, language=CN, stringName=李琰, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 云南民族大学电气信息工程学院云南省高校电力信息物理融合系统重点实验室, 昆明 650504, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1217860108982337665, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, xref=1, ext=[AuthorCompanyExt(id=1217860108999114882, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, companyId=1217860108982337665, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Key Laboratory of Cyber-Physical Power System of Yunnan Universities, School of Electrical and Information Technology, Yunnan Minzu University, Kunming 650504, China), AuthorCompanyExt(id=1217860109015892100, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, companyId=1217860108982337665, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 云南民族大学电气信息工程学院云南省高校电力信息物理融合系统重点实验室, 昆明 650504)])]), Author(id=1217860111918350642, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, orderNo=3, 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=1217860112182591812, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, authorId=1217860111918350642, language=EN, stringName=Xiao-lan LI, firstName=Xiao-lan, middleName=null, lastName=LI, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 Key Laboratory of Cyber-Physical Power System of Yunnan Universities, School of Electrical and Information Technology, Yunnan Minzu University, Kunming 650504, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1217860112350363985, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, authorId=1217860111918350642, language=CN, stringName=李晓兰, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 云南民族大学电气信息工程学院云南省高校电力信息物理融合系统重点实验室, 昆明 650504, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1217860108982337665, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, xref=1, ext=[AuthorCompanyExt(id=1217860108999114882, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, companyId=1217860108982337665, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Key Laboratory of Cyber-Physical Power System of Yunnan Universities, School of Electrical and Information Technology, Yunnan Minzu University, Kunming 650504, China), AuthorCompanyExt(id=1217860109015892100, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, companyId=1217860108982337665, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 云南民族大学电气信息工程学院云南省高校电力信息物理融合系统重点实验室, 昆明 650504)])])], keywords=[Keyword(id=1217860112702685558, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=EN, orderNo=1, keyword=transient stability), Keyword(id=1217860112799154568, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=EN, orderNo=2, keyword=data-driven), Keyword(id=1217860112887234964, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=EN, orderNo=3, keyword=attention mechanism), Keyword(id=1217860112996286881, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=EN, orderNo=4, keyword=Transformer), Keyword(id=1217860113109533104, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=EN, orderNo=5, keyword=graph attention network), Keyword(id=1217860113247945155, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=EN, orderNo=6, keyword=feature fusion), Keyword(id=1217860113336025555, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=CN, orderNo=1, keyword=暂态稳定性), Keyword(id=1217860113445077473, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=CN, orderNo=2, keyword=数据驱动), Keyword(id=1217860113537352171, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=CN, orderNo=3, keyword=注意力机制), Keyword(id=1217860113654792698, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=CN, orderNo=4, keyword=Transformer), Keyword(id=1217860113860313618, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=CN, orderNo=5, keyword=图注意力网络), Keyword(id=1217860113952588318, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=CN, orderNo=6, keyword=特征融合)], refs=[Reference(id=1217860121598804985, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=https://www.china5e.com/energy/news-334763-1.html, language=null, rfNumber=[1], rfOrder=0, authorNames=中国能源网, journalName=null, refType=null, unstructuredReference=中国能源网. 华中电网事故的经验与教训[EB/OL]. ( 2017-08-09) [2025-01-17]. https://www.china5e.com/energy/news-334763-1.html., articleTitle=华中电网事故的经验与教训, refAbstract=null), Reference(id=1217860121733021700, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=https://www.china5e.com/energy/news-334763-1.html, language=null, rfNumber=[1], rfOrder=1, authorNames=China Energy Network, journalName=null, refType=null, unstructuredReference=China Energy Network. Experience and lessons from the central China power grid accident[EB/OL]. ( 2017-08-09) [2025-01-17]. https://www.china5e.com/energy/news-334763-1.html., articleTitle=Experience and lessons from the central China power grid accident, refAbstract=null), Reference(id=1217860121892405262, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2022, volume=42, issue=8, pageStart=2806, pageEnd=2819, url=null, language=null, rfNumber=[2], rfOrder=2, authorNames=张智刚, 康重庆, journalName=中国电机工程学报, refType=null, unstructuredReference=张智刚, 康重庆. 碳中和目标下构建新型电力系统的挑战与展望[J]. 中国电机工程学报, 2022, 42(8): 2806-2819., articleTitle=碳中和目标下构建新型电力系统的挑战与展望, refAbstract=null), Reference(id=1217860122068566042, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2022, volume=42, issue=8, pageStart=2806, pageEnd=2819, url=null, language=null, rfNumber=[2], rfOrder=3, authorNames=Zhang Zhigang, Kang Chongqing, journalName=Proceedings of the CSEE, refType=null, unstructuredReference=Zhang Zhigang, Kang Chongqing. Challenges and prospects of building a new power system under the carbon neutrality goal[J]. Proceedings of the CSEE, 2022, 42(8): 2806-2819., articleTitle=Challenges and prospects of building a new power system under the carbon neutrality goal, refAbstract=null), Reference(id=1217860122190200866, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=28, pageStart=12177, pageEnd=12187, url=null, language=null, rfNumber=[3], rfOrder=4, authorNames=朱冰, 张龙, 司文佳, journalName=科学技术与工程, refType=null, unstructuredReference=朱冰, 张龙, 司文佳, 等. 跟网型与构网型变流器混合并网的暂态特性分析及故障穿越控制[J]. 科学技术与工程, 2024, 24(28): 12177-12187., articleTitle=跟网型与构网型变流器混合并网的暂态特性分析及故障穿越控制, refAbstract=null), Reference(id=1217860122286669868, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=28, pageStart=12177, pageEnd=12187, url=null, language=null, rfNumber=[3], rfOrder=5, authorNames=Zhu Bing, Zhang Long, Si Wenjia, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Zhu Bing, Zhang Long, Si Wenjia, et al. Analysis of transient characteristics and fault ride-through control of hybrid grid-tied converters with grid-following and grid-forming[J]. Science Technology and Engineering, 2024, 24(28): 12177-12187., articleTitle=Analysis of transient characteristics and fault ride-through control of hybrid grid-tied converters with grid-following and grid-forming, refAbstract=null), Reference(id=1217860122404110393, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2022, volume=22, issue=2, pageStart=556, pageEnd=562, url=null, language=null, rfNumber=[4], rfOrder=6, authorNames=高剑, 毕悦, 杨亚兰, journalName=科学技术与工程, refType=null, unstructuredReference=高剑, 毕悦, 杨亚兰, 等. 光伏电站机电暂态实用化等值建模方法[J]. 科学技术与工程, 2022, 22(2): 556-562., articleTitle=光伏电站机电暂态实用化等值建模方法, refAbstract=null), Reference(id=1217860122534133820, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2022, volume=22, issue=2, pageStart=556, pageEnd=562, url=null, language=null, rfNumber=[4], rfOrder=7, authorNames=Gao Jian, Bi Yue, Yang Yalan, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Gao Jian, Bi Yue, Yang Yalan, et al. Practical equivalent modeling method of photovoltaic power station in electromechanical transient simulation[J]. Science Technology and Engineering, 2022, 22(2): 556-562., articleTitle=Practical equivalent modeling method of photovoltaic power station in electromechanical transient simulation, refAbstract=null), Reference(id=1217860122601242690, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2016, volume=36, issue=5, pageStart=1173, pageEnd=1180, url=null, language=null, rfNumber=[5], rfOrder=8, authorNames=戴远航, 陈磊, 张玮灵, journalName=中国电机工程学报, refType=null, unstructuredReference=戴远航, 陈磊, 张玮灵, 等. 基于多支持向量机综合的电力系统暂态稳定评估[J]. 中国电机工程学报, 2016, 36(5): 1173-1180., articleTitle=基于多支持向量机综合的电力系统暂态稳定评估, refAbstract=null), Reference(id=1217860122697711692, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2016, volume=36, issue=5, pageStart=1173, pageEnd=1180, url=null, language=null, rfNumber=[5], rfOrder=9, authorNames=Dai Yuanhang, Chen Lei, Zhang Weiling, journalName=Proceedings of the CSEE, refType=null, unstructuredReference=Dai Yuanhang, Chen Lei, Zhang Weiling, et al. Multi-support vector machine power system transient stability assessment based on relief algorithm[J]. Proceedings of the CSEE, 2016, 36(5): 1173-1180., articleTitle=Multi-support vector machine power system transient stability assessment based on relief algorithm, refAbstract=null), Reference(id=1217860122810957907, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2022, volume=22, issue=11, pageStart=4367, pageEnd=4374, url=null, language=null, rfNumber=[6], rfOrder=10, authorNames=刘炼, 王强, 陈浩, journalName=科学技术与工程, refType=null, unstructuredReference=刘炼, 王强, 陈浩. 基于改进随机森林的电力系统暂态稳定评估[J]. 科学技术与工程, 2022, 22(11): 4367-4374., articleTitle=基于改进随机森林的电力系统暂态稳定评估, refAbstract=null), Reference(id=1217860122957758554, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2022, volume=22, issue=11, pageStart=4367, pageEnd=4374, url=null, language=null, rfNumber=[6], rfOrder=11, authorNames=Liu Lian, Wang Qiang, Chen Hao, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Liu Lian, Wang Qiang, Chen Hao. Transient stability assessment of power systems based on improved random forest[J]. Science Technology and Engineering, 2022, 22(11): 4367-4374., articleTitle=Transient stability assessment of power systems based on improved random forest, refAbstract=null), Reference(id=1217860123075199076, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2024, volume=55, issue=1, pageStart=73, pageEnd=83, url=null, language=null, rfNumber=[7], rfOrder=12, authorNames=曲莹, 韩肖清, 刘新元, journalName=太原理工大学学报, refType=null, unstructuredReference=曲莹, 韩肖清, 刘新元, 等. 基于多源数据驱动的电力系统暂态稳定性分析方法[J]. 太原理工大学学报, 2024, 55(1): 73-83., articleTitle=基于多源数据驱动的电力系统暂态稳定性分析方法, refAbstract=null), Reference(id=1217860123192639599, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2024, volume=55, issue=1, pageStart=73, pageEnd=83, url=null, language=null, rfNumber=[7], rfOrder=13, authorNames=Qu Ying, Han Xiaoqing, Liu Xinyuan, journalName=Journal of Taiyuan University of Technology, refType=null, unstructuredReference=Qu Ying, Han Xiaoqing, Liu Xinyuan, et al. Analysis method for transient stability of power systems based on multi-source data-driven approach[J]. Journal of Taiyuan University of Technology, 2024, 55(1): 73-83., articleTitle=Analysis method for transient stability of power systems based on multi-source data-driven approach, refAbstract=null), Reference(id=1217860123414937721, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2020, volume=20, issue=1, pageStart=213, pageEnd=217, url=null, language=null, rfNumber=[8], rfOrder=14, authorNames=李向伟, 刘思言, 高昆仑, journalName=科学技术与工程, refType=null, unstructuredReference=李向伟, 刘思言, 高昆仑. 基于差分进化极限学习机的电力系统暂态稳定评估方法[J]. 科学技术与工程, 2020, 20(1): 213-217., articleTitle=基于差分进化极限学习机的电力系统暂态稳定评估方法, refAbstract=null), Reference(id=1217860123549155459, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2020, volume=20, issue=1, pageStart=213, pageEnd=217, url=null, language=null, rfNumber=[8], rfOrder=15, authorNames=Li Xiangwei, Liu Siyan, Gao Kunlun, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Li Xiangwei, Liu Siyan, Gao Kunlun. Power system transient stability assessment based on the differential evolution extreme learning machine[J]. Science Technology and Engineering, 2020, 20(1): 213-217., articleTitle=Power system transient stability assessment based on the differential evolution extreme learning machine, refAbstract=null), Reference(id=1217860123641430160, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2025, volume=45, issue=9, pageStart=3436, pageEnd=3448, url=null, language=null, rfNumber=[9], rfOrder=16, authorNames=周芮, 杨燕, 余娟, journalName=中国电机工程学报, refType=null, unstructuredReference=周芮, 杨燕, 余娟, 等. 适应拓扑变化的数据驱动电力系统暂态主导失稳模式识别方法[J]. 中国电机工程学报, 2025, 45(9): 3436-3448., articleTitle=适应拓扑变化的数据驱动电力系统暂态主导失稳模式识别方法, refAbstract=null), Reference(id=1217860123733704857, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2025, volume=45, issue=9, pageStart=3436, pageEnd=3448, url=null, language=null, rfNumber=[9], rfOrder=17, authorNames=Zhou Rui, Yang Yan, Yu Juan, journalName=IEEE Transactions on Power Engineering, refType=null, unstructuredReference=Zhou Rui, Yang Yan, Yu Juan, et al. Data-driven method for identifying transient dominant instability patterns in power systems adapted to topological changes[J]. IEEE Transactions on Power Engineering, 2025, 45(9): 3436-3448., articleTitle=Data-driven method for identifying transient dominant instability patterns in power systems adapted to topological changes, refAbstract=null), Reference(id=1217860123834368161, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2022, volume=22, issue=4, pageStart=1456, pageEnd=1464, url=null, language=null, rfNumber=[10], rfOrder=18, authorNames=王强, 陈浩, 刘炼, journalName=科学技术与工程, refType=null, unstructuredReference=王强, 陈浩, 刘炼. 基于多层CatBoost的电力系统暂态稳定评估[J]. 科学技术与工程, 2022, 22(4): 1456-1464., articleTitle=基于多层CatBoost的电力系统暂态稳定评估, refAbstract=null), Reference(id=1217860123968585895, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2022, volume=22, issue=4, pageStart=1456, pageEnd=1464, url=null, language=null, rfNumber=[10], rfOrder=19, authorNames=Wang Qiang, Chen Hao, Liu Lian, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Wang Qiang, Chen Hao, Liu Lian. Research on transient stability assessment of power system based on multi-layer CatBoost[J]. Science Technology and Engineering, 2022, 22(4): 1456-1464., articleTitle=Research on transient stability assessment of power system based on multi-layer CatBoost, refAbstract=null), Reference(id=1217860125314957486, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2022, volume=50, issue=11, pageStart=17, pageEnd=24, url=null, language=null, rfNumber=[11], rfOrder=20, authorNames=彭鑫, 刘俊, 刘嘉诚, journalName=智慧电力, refType=null, unstructuredReference=彭鑫, 刘俊, 刘嘉诚, 等. 图像化数据驱动的电力系统暂态稳定性在线评估方法[J]. 智慧电力, 2022, 50(11): 17-24., articleTitle=图像化数据驱动的电力系统暂态稳定性在线评估方法, refAbstract=null), Reference(id=1217860125461758134, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2022, volume=50, issue=11, pageStart=17, pageEnd=24, url=null, language=null, rfNumber=[11], rfOrder=21, authorNames=Peng Xin, Liu Jun, Liu Jiacheng, journalName=Smart Power, refType=null, unstructuredReference=Peng Xin, Liu Jun, Liu Jiacheng, et al. Data-driven online assessment method for transient stability of power systems based on visualization[J]. Smart Power, 2022, 50(11): 17-24., articleTitle=Data-driven online assessment method for transient stability of power systems based on visualization, refAbstract=null), Reference(id=1217860125587587263, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2022, volume=7, issue=3, pageStart=1, pageEnd=16, url=null, language=null, rfNumber=[12], rfOrder=22, authorNames=Cheng S, Yu Z, Liu Y, journalName=Protection and Control of Modern Power Systems, refType=null, unstructuredReference=Cheng S, Yu Z, Liu Y, et al. Power system transient stability assessment based on the multiple paralleled convolutional neural network and gated recurrent unit[J]. Protection and Control of Modern Power Systems, 2022, 7(3): 1-16., articleTitle=Power system transient stability assessment based on the multiple paralleled convolutional neural network and gated recurrent unit, refAbstract=null), Reference(id=1217860125738582215, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2020, volume=20, issue=7, pageStart=2733, pageEnd=2739, url=null, language=null, rfNumber=[13], rfOrder=23, authorNames=李向伟, 刘思言, 高昆仑, journalName=科学技术与工程, refType=null, unstructuredReference=李向伟, 刘思言, 高昆仑. 基于双向长短时记忆网络和卷曲神经网络的电力系统暂态稳定评估[J]. 科学技术与工程, 2020, 20(7): 2733-2739., articleTitle=基于双向长短时记忆网络和卷曲神经网络的电力系统暂态稳定评估, refAbstract=null), Reference(id=1217860125889577167, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2020, volume=20, issue=7, pageStart=2733, pageEnd=2739, url=null, language=null, rfNumber=[13], rfOrder=24, authorNames=Li Xiangwei, Liu Siyan, Gao Kunlun, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Li Xiangwei, Liu Siyan, Gao Kunlun. Power system transient stability assessment based on BiLSTM-ICNN[J]. Science Technology and Engineering, 2020, 20(7): 2733-2739., articleTitle=Power system transient stability assessment based on BiLSTM-ICNN, refAbstract=null), Reference(id=1217860126019600601, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2025, volume=40, issue=2, pageStart=1214, pageEnd=1227, url=null, language=null, rfNumber=[14], rfOrder=25, authorNames=Wang H, Gao F, Chen Q, journalName=IEEE Transactions on Power Systems, refType=null, unstructuredReference=Wang H, Gao F, Chen Q, et al. Instability pattern-guided model updating method for data-driven transient stability assessment[J]. IEEE Transactions on Power Systems, 2025, 40(2): 1214-1227., articleTitle=Instability pattern-guided model updating method for data-driven transient stability assessment, refAbstract=null), Reference(id=1217860126183178463, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2022, volume=null, issue=null, pageStart=170, pageEnd=175, url=null, language=null, rfNumber=[15], rfOrder=26, authorNames=Zhou J, Bu S, journalName=12th IET International Conference on Advances in Power System Control, Operation and Management (APSCOM 2022), Hybrid Conference, refType=null, unstructuredReference=Zhou J, Bu S. TSPM-CNN based power system transient stability assessment using a new data extraction method[C]// 12th IET International Conference on Advances in Power System Control, Operation and Management (APSCOM 2022), Hybrid Conference. Hong Kong: Research Institute for Smart Energy, The Hong Kong Polytechnic University, 2022: 170-175., articleTitle=TSPM-CNN based power system transient stability assessment using a new data extraction method, refAbstract=null), Reference(id=1217860126325784813, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2024, volume=20, issue=7, pageStart=9083, pageEnd=9097, url=null, language=null, rfNumber=[16], rfOrder=27, authorNames=Shao Z, Wang Q, Cao Y, journalName=IEEE Transactions on Industrial Informatics, refType=null, unstructuredReference=Shao Z, Wang Q, Cao Y, et al. A novel data-driven LSTM-SAF model for power systems transient stability assessment[J]. IEEE Transactions on Industrial Informatics, 2024, 20(7): 9083-9097., articleTitle=A novel data-driven LSTM-SAF model for power systems transient stability assessment, refAbstract=null), Reference(id=1217860126464196848, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2024, volume=12, issue=null, pageStart=125095, pageEnd=125107, url=null, language=null, rfNumber=[17], rfOrder=28, authorNames=Umbereen S, Weiss X, Rolander A, journalName=IEEE Access, refType=null, unstructuredReference=Umbereen S, Weiss X, Rolander A, et al. Investigating the performance of MLE & CNN for transient stability assessment in power systems[J]. IEEE Access, 2024, 12: 125095-125107., articleTitle=Investigating the performance of MLE & CNN for transient stability assessment in power systems, refAbstract=null), Reference(id=1217860126585831676, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2023, volume=43, issue=15, pageStart=5745, pageEnd=5759, url=null, language=null, rfNumber=[18], rfOrder=29, authorNames=房佳姝, 刘崇茹, 苏晨博, journalName=中国电机工程学报, refType=null, unstructuredReference=房佳姝, 刘崇茹, 苏晨博, 等. 基于自注意力Transformer编码器的多阶段电力系统暂态稳定评估方法[J]. 中国电机工程学报, 2023, 43(15): 5745-5759., articleTitle=基于自注意力Transformer编码器的多阶段电力系统暂态稳定评估方法, refAbstract=null), Reference(id=1217860126736826629, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2023, volume=43, issue=15, pageStart=5745, pageEnd=5759, url=null, language=null, rfNumber=[18], rfOrder=30, authorNames=Fang Jiashu, Liu Chongru, Su Chenbo, journalName=IEEE Transactions on Power Engineering, refType=null, unstructuredReference=Fang Jiashu, Liu Chongru, Su Chenbo, et al. Multi-stage transient stability assessment method for power systems based on self-attention transformer encoder[J]. IEEE Transactions on Power Engineering, 2023, 43(15): 5745-5759., articleTitle=Multi-stage transient stability assessment method for power systems based on self-attention transformer encoder, refAbstract=null), Reference(id=1217860126858461449, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=777, pageEnd=782, url=null, language=null, rfNumber=[19], rfOrder=31, authorNames=Mahato N K, Dong J, Song C, journalName=2021 6th Asia Conference on Power and Electrical Engineering (ACPEE), refType=null, unstructuredReference=Mahato N K, Dong J, Song C, et al. Electric power system transient stability assessment based on Bi-LSTM attention mechanism[C]// 2021 6th Asia Conference on Power and Electrical Engineering (ACPEE). Chongqing: Engineering Research Center of Energy Electric Power Information Security, 2021: 777-782., articleTitle=Electric power system transient stability assessment based on Bi-LSTM attention mechanism, refAbstract=null), Reference(id=1217860127047205139, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2023, volume=47, issue=7, pageStart=114, pageEnd=122, url=null, language=null, rfNumber=[20], rfOrder=32, authorNames=张亮, 安军, 周毅博, journalName=电力系统自动化, refType=null, unstructuredReference=张亮, 安军, 周毅博. 基于时间卷积和图注意力网络的电力系统暂态稳定评估[J]. 电力系统自动化, 2023, 47(7): 114-122., articleTitle=基于时间卷积和图注意力网络的电力系统暂态稳定评估, refAbstract=null), Reference(id=1217860127185617182, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2023, volume=47, issue=7, pageStart=114, pageEnd=122, url=null, language=null, rfNumber=[20], rfOrder=33, authorNames=Zhang Liang, An Jun, Zhou Yibo, journalName=Automation of Electric Power Systems, refType=null, unstructuredReference=Zhang Liang, An Jun, Zhou Yibo. Transient stability assessment of power systems based on temporal convolution and graph attention networks[J]. Automation of Electric Power Systems, 2023, 47(7): 114-122., articleTitle=Transient stability assessment of power systems based on temporal convolution and graph attention networks, refAbstract=null), Reference(id=1217860127345000739, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[21], rfOrder=34, authorNames=Velikovi P, Cucurull G, Casanova A, journalName=6th International Conference on Learning Representations, refType=null, unstructuredReference=Velikovi P, Cucurull G, Casanova A, et al. Graph attention networks[C]// 6th International Conference on Learning Representations. Cambridge: University of Cambridge, 2018., articleTitle=Graph attention networks, refAbstract=null), Reference(id=1217860127449858345, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=null, pageStart=1710, pageEnd=null, url=null, language=null, rfNumber=[22], rfOrder=35, authorNames=Vaswani A, Shazeer N, Parmar A, journalName=ArXiv, refType=null, unstructuredReference=Vaswani A, Shazeer N, Parmar A, et al. Attention is all you need[J]. ArXiv, 2017: 1710.10903., articleTitle=Attention is all you need, refAbstract=null), Reference(id=1217860127537938736, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2015, volume=39, issue=1, pageStart=73, pageEnd=80, url=null, language=null, rfNumber=[23], rfOrder=36, authorNames=段刚, 严亚勤, 谢晓冬, journalName=电力系统自动化, refType=null, unstructuredReference=段刚, 严亚勤, 谢晓冬, 等. 广域相量测量技术发展现状与展望[J]. 电力系统自动化, 2015, 39(1): 73-80., articleTitle=广域相量测量技术发展现状与展望, refAbstract=null), Reference(id=1217860127655379254, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2015, volume=39, issue=1, pageStart=73, pageEnd=80, url=null, language=null, rfNumber=[23], rfOrder=37, authorNames=Duan Gang, Yan Yaqin, Xie Xiaodong, journalName=Automation of Electric Power Systems, refType=null, unstructuredReference=Duan Gang, Yan Yaqin, Xie Xiaodong, et al. Current status and prospects of wide-area phasor measurement technology[J]. Automation of Electric Power Systems, 2015, 39(1): 73-80., articleTitle=Current status and prospects of wide-area phasor measurement technology, refAbstract=null), Reference(id=1217860127806374206, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2024, volume=19, issue=4, pageStart=246, pageEnd=254, url=null, language=null, rfNumber=[24], rfOrder=38, authorNames=周宇, 肖健梅, 王锡淮, journalName=电气工程学报, refType=null, unstructuredReference=周宇, 肖健梅, 王锡淮. 基于GCN和HGP-SL的电力系统暂态稳定评估[J]. 电气工程学报, 2024, 19(4): 246-254., articleTitle=基于GCN和HGP-SL的电力系统暂态稳定评估, refAbstract=null), Reference(id=1217860127898648902, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2024, volume=19, issue=4, pageStart=246, pageEnd=254, url=null, language=null, rfNumber=[24], rfOrder=39, authorNames=Zhou Yu, Xiao Jianmei, Wang Xihuai, journalName=Journal of Electrical Engineering, refType=null, unstructuredReference=Zhou Yu, Xiao Jianmei, Wang Xihuai. Transient stability assessment of power systems based on GCN and HGP-SL[J]. Journal of Electrical Engineering, 2024, 19(4): 246-254., articleTitle=Transient stability assessment of power systems based on GCN and HGP-SL, refAbstract=null), Reference(id=1217860127999312203, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2021, volume=41, issue=7, pageStart=2341, pageEnd=2350, url=null, language=null, rfNumber=[25], rfOrder=40, authorNames=王铮澄, 周艳真, 郭庆来, journalName=中国电机工程学报, refType=null, unstructuredReference=王铮澄, 周艳真, 郭庆来, 等. 考虑电力系统拓扑变化的消息传递图神经网络暂态稳定评估[J]. 中国电机工程学报, 2021, 41(7): 2341-2350., articleTitle=考虑电力系统拓扑变化的消息传递图神经网络暂态稳定评估, refAbstract=null), Reference(id=1217860128087392592, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, doi=null, pmid=null, pmcid=null, year=2021, volume=41, issue=7, pageStart=2341, pageEnd=2350, url=null, language=null, rfNumber=[25], rfOrder=41, authorNames=Wang Zhengcheng, Zhou Yanzhen, Guo Qinglai, journalName=IEEE Transactions on Power Engineering, refType=null, unstructuredReference=Wang Zhengcheng, Zhou Yanzhen, Guo Qinglai, et al. Transient stability assessment of power systems considering topological changes using message passing graph neural networks[J]. IEEE Transactions on Power Engineering, 2021, 41(7): 2341-2350., articleTitle=Transient stability assessment of power systems considering topological changes using message passing graph neural networks, refAbstract=null)], funds=[Fund(id=1217860121099682772, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, awardId=62062068, language=CN, fundingSource=国家自然科学基金(62062068), fundOrder=null, country=null), Fund(id=1217860121238094813, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, awardId=202305AC160077, language=CN, fundingSource=云南省中青年学术和技术带头人后备人才项目(202305AC160077), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1217860108982337665, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, xref=1, ext=[AuthorCompanyExt(id=1217860108999114882, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, companyId=1217860108982337665, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Key Laboratory of Cyber-Physical Power System of Yunnan Universities, School of Electrical and Information Technology, Yunnan Minzu University, Kunming 650504, China), AuthorCompanyExt(id=1217860109015892100, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, companyId=1217860108982337665, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 云南民族大学电气信息工程学院云南省高校电力信息物理融合系统重点实验室, 昆明 650504)]), AuthorCompany(id=1217860109124944015, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, xref=2, ext=[AuthorCompanyExt(id=1217860109145915539, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, companyId=1217860109124944015, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Yunnan Key Laboratory of Unmanned Autonomous System, Kunming 650504, China), AuthorCompanyExt(id=1217860109154304148, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, companyId=1217860109124944015, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 云南省无人自主系统重点实验室, 昆明 650504)])], figs=[ArticleFig(id=1217860114237801028, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=EN, label=Fig.1, caption=Graph attention mechanism, figureFileSmall=u34AFOXlW5+tiI4s0BC3gA==, figureFileBig=5oQLZkjHzd1aO0hGxqoq7A==, tableContent=null), ArticleFig(id=1217860114430739039, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=CN, label=图1, caption=图注意力机制, figureFileSmall=u34AFOXlW5+tiI4s0BC3gA==, figureFileBig=5oQLZkjHzd1aO0hGxqoq7A==, tableContent=null), ArticleFig(id=1217860114581734003, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=EN, label=Fig.2, caption=Multi-head graph attention aggregation, figureFileSmall=phDLC6wc7I4C+7X6g52wTg==, figureFileBig=ASPlq1l5q/82jqJqwdrrfQ==, tableContent=null), ArticleFig(id=1217860114745311873, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=CN, label=图2, caption=多头图注意力聚合, figureFileSmall=phDLC6wc7I4C+7X6g52wTg==, figureFileBig=ASPlq1l5q/82jqJqwdrrfQ==, tableContent=null), ArticleFig(id=1217860116011991697, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=EN, label=Fig.3, caption=Transformer architecture, figureFileSmall=TyZZ1d0q+7Z4dIIuUBnI3A==, figureFileBig=p6ner7fBBsOziD5X7w0wDw==, tableContent=null), ArticleFig(id=1217860116175569569, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=CN, label=图3, caption=Transformer架构, figureFileSmall=TyZZ1d0q+7Z4dIIuUBnI3A==, figureFileBig=p6ner7fBBsOziD5X7w0wDw==, tableContent=null), ArticleFig(id=1217860116313981612, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=EN, label=Fig.4, caption=Parallel feature fusion of Transformer-GAT, figureFileSmall=eH3t0ML52kHjboEfTqH09w==, figureFileBig=tk/DNqDK3LUEr0oYfsIcZg==, tableContent=null), ArticleFig(id=1217860116473365175, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=CN, label=图4, caption=Transformer-GAT并行特征融合, figureFileSmall=eH3t0ML52kHjboEfTqH09w==, figureFileBig=tk/DNqDK3LUEr0oYfsIcZg==, tableContent=null), ArticleFig(id=1217860116615971520, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=EN, label=Fig.5, caption=TSA flowchart, figureFileSmall=GufkEd9GSn0B8xnctG+eXg==, figureFileBig=X1V3l3qZDMowbLdXeXsrog==, tableContent=null), ArticleFig(id=1217860116758577876, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=CN, label=图5, caption=TSA流程图, figureFileSmall=GufkEd9GSn0B8xnctG+eXg==, figureFileBig=X1V3l3qZDMowbLdXeXsrog==, tableContent=null), ArticleFig(id=1217860116876018404, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=EN, label=Fig.6, caption=IEEE 39-bus system topology diagram, figureFileSmall=cSWjH5kcgPxyPU01TAaWTA==, figureFileBig=7ixbJog5wM5vv+BsKK8gfw==, tableContent=null), ArticleFig(id=1217860117022819055, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=CN, label=图6, caption=IEEE39节点系统拓扑图, figureFileSmall=cSWjH5kcgPxyPU01TAaWTA==, figureFileBig=7ixbJog5wM5vv+BsKK8gfw==, tableContent=null), ArticleFig(id=1217860117173814011, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=EN, label=Fig.7, caption=Serial configuration of Transformer-GAT, figureFileSmall=tthZ1urwozr5CuSjfvUNWQ==, figureFileBig=LxBDae3mSdW7Wbb5kLKLEg==, tableContent=null), ArticleFig(id=1217860117308031748, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=CN, label=图7, caption=Transformer和GAT串行方式, figureFileSmall=tthZ1urwozr5CuSjfvUNWQ==, figureFileBig=LxBDae3mSdW7Wbb5kLKLEg==, tableContent=null), ArticleFig(id=1217860117417083662, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=EN, label=Fig.8, caption=Confusion matrix, figureFileSmall=tYZ59Q5ei2boYxkP80LKPw==, figureFileBig=9IeOG+xhJ/ciWsVYFYZKvQ==, tableContent=null), ArticleFig(id=1217860117572272926, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=CN, label=图8, caption=混淆矩阵, figureFileSmall=tYZ59Q5ei2boYxkP80LKPw==, figureFileBig=9IeOG+xhJ/ciWsVYFYZKvQ==, tableContent=null), ArticleFig(id=1217860117693907753, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=EN, label=Fig.9, caption=Comparison of model convergence, figureFileSmall=LwoH0DD1PSTSfedNDCtgJg==, figureFileBig=7Z9DrBl8oIM4o2GYvv5N8A==, tableContent=null), ArticleFig(id=1217860117849097014, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=CN, label=图9, caption=模型收敛性对比, figureFileSmall=LwoH0DD1PSTSfedNDCtgJg==, figureFileBig=7Z9DrBl8oIM4o2GYvv5N8A==, tableContent=null), ArticleFig(id=1217860118104949573, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=EN, label=Fig.10, caption=Prediction accuracy of six models, figureFileSmall=7kxt0NBJRkKnK2yQhzxzxQ==, figureFileBig=T+4LNAiKJs9gk7bBexu0jA==, tableContent=null), ArticleFig(id=1217860118293693269, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=CN, label=图10, caption=6种模型的预测准确率, figureFileSmall=7kxt0NBJRkKnK2yQhzxzxQ==, figureFileBig=T+4LNAiKJs9gk7bBexu0jA==, tableContent=null), ArticleFig(id=1217860118570517348, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=EN, label=Fig.11, caption=Comparison of model performance metrics after adding noise, figureFileSmall=qTDH+oaOoChB1/3poYIhOQ==, figureFileBig=m6Nq50v0lmkn2YVf0zTOHA==, tableContent=null), ArticleFig(id=1217860118717318004, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=CN, label=图11, caption=加入噪声后模型性能指标对比, figureFileSmall=qTDH+oaOoChB1/3poYIhOQ==, figureFileBig=m6Nq50v0lmkn2YVf0zTOHA==, tableContent=null), ArticleFig(id=1217860118822175612, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=EN, label=Table 1, caption=

Transformer-GAT parallel fusion algorithm

, figureFileSmall=null, figureFileBig=null, tableContent=
初始化:数据集节点特征、边索引和标签(node_features, edge_index,labels),初始化Transformer、GAT参数
1.准备输入数据,包括节点特征和邻接矩阵:Data(x=node_features, edge_index=edge_index)
2.定义Transformer和GAT模型结构:TransformerLayer()和GATConv()
3.用Transformer处理节点特征数据,提取时间序列特征。用GAT处理节点特征数据,提取图结构特征:transformer(node_features)、gat(node_features, edge_index)
4.将Transformer和GAT的输出进行加权方法融合,fusion(transformer, gat)
5.损失计算和反向传播:loss(output,labels)、loss.backward()
6.将融合后的特征传递给分类器进行最终预测:classifier(fusion_out)
7.模型评估、保存模型参数
), ArticleFig(id=1217860118935421826, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=CN, label=表1, caption=

Transfoemer-GAT并行融合模型

, figureFileSmall=null, figureFileBig=null, tableContent=
初始化:数据集节点特征、边索引和标签(node_features, edge_index,labels),初始化Transformer、GAT参数
1.准备输入数据,包括节点特征和邻接矩阵:Data(x=node_features, edge_index=edge_index)
2.定义Transformer和GAT模型结构:TransformerLayer()和GATConv()
3.用Transformer处理节点特征数据,提取时间序列特征。用GAT处理节点特征数据,提取图结构特征:transformer(node_features)、gat(node_features, edge_index)
4.将Transformer和GAT的输出进行加权方法融合,fusion(transformer, gat)
5.损失计算和反向传播:loss(output,labels)、loss.backward()
6.将融合后的特征传递给分类器进行最终预测:classifier(fusion_out)
7.模型评估、保存模型参数
), ArticleFig(id=1217860119078028167, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=EN, label=Table 2, caption=

Confusion matrix

, figureFileSmall=null, figureFileBig=null, tableContent=
评估结果 真实标签
稳定 失稳
稳定 TP FP
失稳 FN TN
), ArticleFig(id=1217860120034329490, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=CN, label=表2, caption=

混淆矩阵

, figureFileSmall=null, figureFileBig=null, tableContent=
评估结果 真实标签
稳定 失稳
稳定 TP FP
失稳 FN TN
), ArticleFig(id=1217860120227267487, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=EN, label=Table 3, caption=

Evaluation results of two models

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 A/% F1/% TNR/% TPR/%
Transformer-GAT
并行特征融合
98.73 98.01 97.73 98.28
Transformer和GAT串行 96.27 96.23 95.38 96.55
), ArticleFig(id=1217860120436982702, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=CN, label=表3, caption=

两种模型评估结果

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 A/% F1/% TNR/% TPR/%
Transformer-GAT
并行特征融合
98.73 98.01 97.73 98.28
Transformer和GAT串行 96.27 96.23 95.38 96.55
), ArticleFig(id=1217860120621532089, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=EN, label=Table 4, caption=

Evaluation results of model approaches

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 A/% F1/% TNR/% TPR/%
CNN 88.20 87.34 88.41 86.34
RNN 90.70 89.05 90.21 91.27
GCN 91.98 90.36 92.08 91.13
Transformer 94.33 93.69 94.32 93.97
GAT 94.65 94.95 93.14 94.13
Transformer-GAT
并行特征融合
98.73 98.01 97.73 98.28
), ArticleFig(id=1217860120814470081, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789891908391845, language=CN, label=表4, caption=

模型方案评估结果

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 A/% F1/% TNR/% TPR/%
CNN 88.20 87.34 88.41 86.34
RNN 90.70 89.05 90.21 91.27
GCN 91.98 90.36 92.08 91.13
Transformer 94.33 93.69 94.32 93.97
GAT 94.65 94.95 93.14 94.13
Transformer-GAT
并行特征融合
98.73 98.01 97.73 98.28
)], attaches=null, journal=Journal(id=1146119176004939786, delFlag=0, nameCn=科学技术与工程, nameEn=Science Technology and Engineering, nameHistory1=null, nameHistory2=null, issn=1671-1815, eissn=, cn=11-4688/T, coden=null, periodic=4, language=CN, oaType=是, 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=UKU/O7GSka5polgCTkbIIw==, journalPrice=null, startedYear=null, abbrevIsoEn=Sci Technol Eng, journalRemark=null, publicationField=null, createdTime=null, updatedTime=1754445529766, createdBy=null, updatedBy=13701087609, firstLetterCn=S, firstLetterEn=S, subjectCode=Natural Sciences, subjectName=自然科学, subjectCodeEn=Natural Sciences, subjectNameEn=null, picCn=UKU/O7GSka5polgCTkbIIw==, picEn=5hwlULoNwcbj3xUmVi9MAQ==, jcr=null, cjcr=null, exts=[JournalExt(id=1159791870395564357, language=CN, name=科学技术与工程, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.stae.com.cn/jsygc/home, createdTime=1754445529793, updatedTime=1754445529793, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=http://www.stae.com.cn/jsygc/site/menus/20090429150146001, submissionAuthorUrl=http://www.stae.com.cn/jsygc/author/login, submissionEditorUrl=http://www.stae.com.cn/jsygc/editor/login, submissionReviewUrl=http://www.stae.com.cn/jsygc/reviewer/login, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1159791870441701702, language=EN, name=Science Technology and Engineering, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.stae.com.cn/jsygc/home, createdTime=1754445529804, updatedTime=1754445529804, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=http://www.stae.com.cn/jsygc/author/login, submissionEditorUrl=http://www.stae.com.cn/jsygc/editor/login, submissionReviewUrl=http://www.stae.com.cn/jsygc/reviewer/login, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1146123166801305609, websiteList=[Website(id=1148243202391400884, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146123166801305609, 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/kxjsygc/CN, language=CN, createTime=1751692112777, createBy=18614031015, updateTime=1753520965431, updateBy=18614031015, name=科学技术与工程-中文站点, tplId=1146099689490845704, title=科学技术与工程, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1148622798802673703, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=articleTextType, value=kx, createTime=1751782615614, updateTime=1751782615614, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798781702180, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=banner, value=null, createTime=1751782615609, updateTime=1751782615609, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798769119267, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=j86gbwi+p0Idkyl5SzIlmQ==, createTime=1751782615606, updateTime=1751782615606, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798794285094, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1751782615612, updateTime=1751782615612, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798790090789, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1751782615611, updateTime=1751782615611, creator=18614031015, updator=18614031015)]), Website(id=1155914124811976731, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146123166801305609, 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/kxjsygc/EN, language=EN, createTime=1753521003206, createBy=18614031015, updateTime=1753521003206, updateBy=18614031015, name=科学技术与工程-英文站点, tplId=1146101810881728533, title=Science Technology and Engineering, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1155914371227308235, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=articleTextType, value=kx, createTime=1753521061952, updateTime=1753521061952, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371210531016, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=banner, value=null, createTime=1753521061947, updateTime=1753521061947, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371202142407, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=j86gbwi+p0Idkyl5SzIlmQ==, createTime=1753521061945, updateTime=1753521061945, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371223113930, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1753521061950, updateTime=1753521061950, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371218919625, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1753521061949, updateTime=1753521061949, creator=18614031015, updator=18614031015)])], journalTitle=科学技术与工程, weixinUrl=null, journalUrl=null, iacademicId=null, status=0, seqNo=null, journalTitleEn=Science Technology and Engineering, journalPhotoCn=UKU/O7GSka5polgCTkbIIw==, journalPhotoEn=5hwlULoNwcbj3xUmVi9MAQ==, journalFirstLetter=S, 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=null, provinceCode=null, provinceName=null, collectFlag=false), detailUrlCn=https://castjournals.cast.org.cn/joweb/kxjsygc/CN/10.12404/j.issn.1671-1815.2407675, detailUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/10.12404/j.issn.1671-1815.2407675, pdfUrlCn=https://castjournals.cast.org.cn/joweb/kxjsygc/CN/PDF/10.12404/j.issn.1671-1815.2407675, pdfUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/PDF/10.12404/j.issn.1671-1815.2407675, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于Transformer-GAT并行特征融合的数据驱动的电力系统暂态稳定评估
收藏切换
PDF下载
侯一帆 1 , 徐天奇 1, 2, * , 李琰 1 , 李晓兰 1
科学技术与工程 | 论文·电工技术 2025,25(21): 8945-8954
收起
收藏切换
科学技术与工程 | 论文·电工技术 2025, 25(21): 8945-8954
基于Transformer-GAT并行特征融合的数据驱动的电力系统暂态稳定评估
全屏
侯一帆1 , 徐天奇1, 2, * , 李琰1, 李晓兰1
作者信息
  • 1 云南民族大学电气信息工程学院云南省高校电力信息物理融合系统重点实验室, 昆明 650504
  • 2 云南省无人自主系统重点实验室, 昆明 650504
  • 侯一帆(2000—),男,汉族,甘肃庆阳人,硕士研究生。研究方向:数据驱动电力系统。E-mail:

通讯作者:

* 徐天奇(1978—),男,汉族,云南禄丰人,博士,教授。研究方向:韧性电网、新能源发电并网、电力信息物理系统。E-mail:
Data-driven Transient Stability Assessment of Power Systems Based on Transformer-GAT Parallel Feature Fusion
Yi-fan HOU1 , Tian-qi XU1, 2, * , Yan LI1, Xiao-lan LI1
Affiliations
  • 1 Key Laboratory of Cyber-Physical Power System of Yunnan Universities, School of Electrical and Information Technology, Yunnan Minzu University, Kunming 650504, China
  • 2 Yunnan Key Laboratory of Unmanned Autonomous System, Kunming 650504, China
出版时间: 2025-07-28 doi: 10.12404/j.issn.1671-1815.2407675
文章导航
收藏切换

随着电力系统的快速发展,大规模新能源并网及源网荷储协同优化增加了电力电子设备的比重,使得电网稳定性,特别是暂态稳定性评估,变得尤为重要。针对传统方法在拓扑结构考虑不足的问题,提出了一种基于Transformer-图注意力网络(graph attention network,GAT)并行特征融合的深度学习方法,用于电力系统暂态稳定性评估。以母线电压幅值、相角及拓扑邻接矩阵作为输入特征,利用西门子仿真软件PSS/E生成批量数据,并通过Transformer和GAT并行提取特征,采用注意力机制进行加权融合。与其他方法的对比结果表明,该方法在IEEE 39节点系统中模拟了不同负荷条件和故障工况,结果表明评估精度和鲁棒性较高,能够有效提升电力系统的安全稳定性。

暂态稳定性  /  数据驱动  /  注意力机制  /  Transformer  /  图注意力网络  /  特征融合

With the rapid development of the power system, the large-scale integration of new energy into the grid and the coordinated optimization of source-grid-load-storage have increased the proportion of power electronic equipment, making the stability of the power grid, especially the assessment of transient stability, particularly important. Aiming at the problem of insufficient consideration of topological structure in traditional methods, a deep learning method based on Transformer-graph attention network(GAT) parallel feature fusion was proposed for the transient stability evaluation of power systems. The busbar voltage amplitude, phase angle and topological adjacentation matrix were taken as input features. Batch data were generated using the Siemens simulation software PSS/E, and features were extracted in parallel through Transformer and GAT. Weighted fusion was carried out using the attention mechanism. The comparison results with other methods show that this method simulates different load conditions and fault conditions in the IEEE 39-node system. The results indicate that the evaluation accuracy and robustness are relatively high, and it can effectively improve the safety and stability of the power system.

transient stability  /  data-driven  /  attention mechanism  /  Transformer  /  graph attention network  /  feature fusion
侯一帆, 徐天奇, 李琰, 李晓兰. 基于Transformer-GAT并行特征融合的数据驱动的电力系统暂态稳定评估. 科学技术与工程, 2025 , 25 (21) : 8945 -8954 . DOI: 10.12404/j.issn.1671-1815.2407675
Yi-fan HOU, Tian-qi XU, Yan LI, Xiao-lan LI. Data-driven Transient Stability Assessment of Power Systems Based on Transformer-GAT Parallel Feature Fusion[J]. Science Technology and Engineering, 2025 , 25 (21) : 8945 -8954 . DOI: 10.12404/j.issn.1671-1815.2407675
随着全球气候变化和环境问题加剧,实现“碳达峰、碳中和”已成为各国的紧迫任务。并且随着中国电力系统规模的不断扩大以及可再生能源,特别是风能和太阳能的快速发展,电力系统的运行模式变得日益复杂。在这种背景下,传统的稳定性评估方法面临着许多挑战,如动态负荷波动、可再生能源的间歇性和不可预测性。这使得电力系统暂态稳定性评估不仅在技术上要求更高的精度,而且在实际应用中也需要更加灵活和高效的解决方案。2006年7月1日,华中电网系统功率振荡[1]最终导致系统失稳震荡,期间频率最低为49.11 Hz,华中东部电网与川渝电网解列,华中电网与西北电网直流闭锁、与华北电网解列,可见对系统的失稳情况及时进行评估不可或缺。同时,在此背景下,中国电力系统向源网荷储[2]协同运行发展,新能源并网和电力电子器件的广泛应用使得暂态稳定性面临挑战,因此评估电力系统暂态稳定性尤为重要。传统的时域仿真法[3]和直接法因复杂的电网拓扑和建模[4]难题已无法满足需求。当前,更具时效性和可靠性的是数据驱动的暂态稳定性评估,早期采用机器学习如支持向量机(support vector machine,SVM)[5]k-最近邻(k-nearest neighbors,kNN)、随机森林(random forest,RF)[6]等,但因特征提取能力有限,现已逐渐转向深度学习驱动的评估方法。
文献[7]的卷积神经网络(convolutional neural network,CNN)暂态稳定性分析模型虽然采用了并行和串行融合,但忽略了电力系统网络的拓扑结构,缺乏时空特征提取,导致重要信息丢失。文献[8]采用主元分析和序列浮动后向算法筛选特征,并利用差分进化优化的极限学习机(extreme learning machine,ELM)进行暂态稳定评估,却未考虑拓扑信息。文献[9]基于K均值聚类(K-means clustering)聚类和多随机卷积核变换,灵活适应拓扑变化并高效提取时空特征。文献[10]提出了一种基于多层CatBoost的暂态稳定评估方法,利用mRMR特征选择和多数投票法,该方法在IEEE 39节点系统中具有高精度和良好的泛化能力,电力系统的空间结构并未考虑。文献[11]通过二维主成分分析(two-dimensional principal component analysis,2D-PCA)降维时间序列图像并采用CNN进行稳定性预测,但未考虑拓扑结构影响。文献[12]提出的CNN + GRU(gated recurrent unit)模型能够提取短期和长期特征,但忽视了空间特征。文献[13]提出了一种融合改进卷积神经网络(improved convolutional neural network, ICNN)与双向长短时记忆网络(bidirectional long short term memory network, BiLSTM)的暂态稳定评估方法,结合正则化和Dropout防止过拟合,但其特征量只考虑到了时序特征。文献[14]提出的Transformer编码器模型仅关注时序特征。文献[15]提出的TSPM(transient stability program management)-CNN方法有效提取时间序列和特征间相关性,但同样只考虑时序特征。文献[16]的LSTM-SAF(long short-term memory with self-attention fusion)模型通过自注意机制提升了准确性,但忽视了空间信息。文献[17]基于CNN和极大似然估计(maximum likelihood estimation,MLE)方法,仅利用测量的时间序列数据进行分析。文献[18]构建了基于Transformer编码器的电力系统暂态稳定评估方法,使模型快速捕获电力系统前后时刻间的全局状态依赖关系,故障识别准确率较高。
因此现针对以往研究考虑拓扑结构不足,不能同时考虑时间和空间的特征信息,提出一种基于Transformer-GAT并行特征融合的电力系统暂态稳定评估方法,能够有效捕捉输入序列中远距离时刻之间的依赖关系,而不受传统回归神经网络(recurrent neural network,RNN)或Bi-LSTM[19]的序列长度限制,图注意力网络(graph attention network,GAT)的空间结构信息捕捉能力突出[20],该方法结合Transformer的时序特征捕捉能力和GAT的空间结构建模优势,有效处理电力系统的时空数据,提升评估准确性和鲁棒性。Transformer通过注意力机制捕捉时间序列中的全局依赖,适用于电力系统动态响应的长时依赖分析。GAT则利用自注意力机制捕捉电力系统拓扑结构中的节点关系及其相互影响,从而实现时空特征的联合建模。基于Transformer-GAT并行特征融合的方法在不同电力系统条件下有效评估暂态稳定性,避免类似于华中电网失稳的情况再次发生。为验证其可应用性,选取典型电力系统案例进行测试。最终在IEEE-39节点系统中来拟合中国电力系统的故障类型和负荷水平验证所提方法的有效性。
电力系统网络可以把母线抽象成节点,传输线路抽象成边,呈现出一种网状的数据结构,也就是非欧几里得结构,而图网络专门用来处理非欧几里得结构的网络拓扑图。因此电力系统网络就可以描述成G(X,A),X为每个节点的特征,A为邻接矩阵,用来表示网络结构拓扑图。
A ( p , q ) = 1 , p , q A ( p , q ) = 0 , p , q
而在GAT[21]中通过引入注意力机制,为每条边计算动态权重,从而调整邻居节点对中心节点的影响。尽管邻接矩阵仍是输入的一部分,但GAT不再依赖固定的邻接权重,而是基于注意力系数聚合节点特征。邻接矩阵定义了邻居节点集合,只有在矩阵中有边连接的节点才会被考虑。因此,邻接矩阵的信息在聚合过程中被间接利用。图注意力层输入的是一组节点特征x={x1,x2,…,xN},xi∈RF,其中N为节点数,F为每个节点的特征。这层会产生新的节点特征x'={x'1,x'2,…,x'N,},x'i∈RF' 作为输出。为了更好提取特征,需要将输入特征转换成更高维的特征,至少需要一个可学习的线性变换层,每个节点的线性变换参数是共享的,该变换由权重矩阵W∈RF'×F参数化,之后在节点上进行自注意力-共享机制:RF'×RF'→R计算注意力系数,表达式为
eij=a(Wxi,Wxj)
式(2)表明节点j的特征对节点i的重要性。然后用softmax函数对在所有j的选择上进行归一化,表达式为
$ \alpha_{i j}=\operatorname{softmax}_{j}\left(e_{i j}\right)=\frac{\exp \left(e_{i j}\right)}{\sum_{k \in N_{i}} \exp \left(e_{i k}\right)} $
注意力机制a是一个单层前馈神经网络,由一个权重向量a∈R2F'参数化,并应用了 LeakyReLU非线性激活函数。完全展开后,注意力机制(如图1所示)计算的系数可以表示为
$ \alpha_{i j}=\frac{\exp \left\{\operatorname{LeakyReLU}\left[\boldsymbol{a}^{\mathrm{T}}\left(\boldsymbol{W} \boldsymbol{x}_{i} \| \boldsymbol{W} \boldsymbol{x}_{j}\right)\right]\right\}}{\sum_{k \in N_{i}} \exp \left\{\operatorname{LeakyReLU}\left[\boldsymbol{a}^{\mathrm{T}}\left(\boldsymbol{W} \boldsymbol{x}_{i} \| \boldsymbol{W} \boldsymbol{x}_{j}\right)\right]\right\}} $
式(4)中:上角标T表示转置;‖表示拼接操作。
得到归一化的注意力系数,然后计算对应特征的线性组合,以作为每个节点的输出特征,即
$ \boldsymbol{x}_{i}=\sigma\left(\sum_{j \in N_{i}} \alpha_{i j} \boldsymbol{W} \boldsymbol{x}_{j}\right) $
最终在预测层上通过多头注意力机制进行聚合过程,如图2所示,再用softmax进行分类评估。
在电力系统中,电气量是具有时间序列的一种关系,下一时刻的电压相角与上一时刻密切相连,因此不能独立看待。在暂态过程中,例如在发生故障、切换负荷或发电机动作时,电压相角会迅速变化。通过时间序列分析,可以捕捉电压相角的变化趋势,进而评估电力系统的暂态稳定性和响应特性,而Transformer具有很强的序列关系处理能力。
Transformer[22]采用的架构,在编码器和解码器中分别使用了堆叠的自注意力机制和逐点的全连接层,如图3所示的左半部分和右半部分所示。
在这种结构中,编码器可以将电气特征量表示的输入序列(x1,x2,…,xn)映射为连续的序列z=(z1,z2,…,zn)。在给定z的情况下,解码器会逐步生成特征量的输出序列(y1,y2,…,ym)。
编码器:编码器由相同的层组成。每一层包含两个子层。第一个子层是多头自注意力机制,第二个子层是逐点的全连接前馈网络。在每个子层周围使用了残差连接,并在其后进行层归一化,每个子层的输出为LayerNorm[x+Sublayer(x)],模型中的所有子层以及嵌入层都产生相同维度的输出。
解码器由与编码器相同的层组成,增加了一个子层:在编码器输出上执行多头注意力机制。每个子层通过残差连接和层归一化处理,为了防止某个位置关注后续位置,解码器中的自注意力子层引入了屏蔽机制。结合输出嵌入偏移的位置,确保每个位置的预测只依赖于先前的已知输出。
在编码器和解码器堆栈的底部将“位置编码”添加到输入嵌入中使模型能够利用电气量序列的顺序信息。位置编码与嵌入的维度dmodel相同,然后相加。
位置编码是通过使用不同频率的正弦和余弦函数来表示,即
PE(pos,2i)=sin(pos/10 00 0 2 i / d m o d e l)
PE(pos,2i+1)=cos(pos/10 00 0 2 i / d m o d e l)
式中:pos为位置;i为维度。
位置编码的每个维度对应一个正弦波。它可以让模型更容易通过相对位置来学习注意力机制。
电力系统暂态稳定性评估需要捕捉故障后的动态行为。结合Transformer和GAT的并行注意力机制是一种提升模型性能的有效方法。Transformer擅长处理时间序列,通过多头自注意力机制捕捉电压、电流、相角等关键时序数据中的全局依赖,特别是在故障时识别系统的动态变化。GAT利用电力系统的拓扑结构,通过图注意力机制学习节点间的依赖,聚合邻居节点特征,精准判断故障对不同节点的影响。再通过注意力权重对值进行加权求和的并行融合方式,如式(8)所示,并进行全局平均池化。模型结合时间序列与拓扑特征,能够更全面地捕捉时空信息,从而提升暂态稳定性评估的准确性与效率。模型结构如图4所示。
GAT和Transformer的输出进行加权融合,表达式为
Output=βH'+(1-β)Attention(Q,K,V)
式(8)中:H'为GAT提取的特征;QKV分别为Transformer查询、键和值的特征输出。
选用PyTorch框架来研究电力系统暂态稳定性评估是因为PyTorch提供了一个直观且易于使用的编程接口,使得深度学习模型的构建非常灵活,并且PyTorch使用动态计算图(dynamic computation graphs),可以在每一步训练中即时构建网络结构。同时可以利用GPU加速计算,极大地提高了模型训练和推理的速度。这在需要处理大量电力系统数据(如时序电压和相角数据)时尤为重要,可以大幅度减少模型的训练时间。
Transfoemer-GAT并行融合模型过程对暂态数据集进行训练,如表1所示,循环此过程直至模型性能满足要求。
Transformer模型层数选取4层,注意力头数设置为4个,隐藏层维度: 设置为128,前馈网络大小设为隐藏层维度的2倍。图注意力网络图选用2层的图注意力网络。层数过多会导致图卷积中的过平滑问题,使得节点特征趋于一致,影响区分度。注意力头数:类似于Transformer,选择4个头以增强模型的多样性。节点特征维度为2。图注意力机制的激活函数使用Leaky ReLU作为激活函数可以增强模型的非线性表示能力。特征融合模块融合方式: 采用注意力机制方式来融合Transformer和GAT提取的特征。注意力机制能平衡不同模型特征的影响,能够保留更多信息。优化器使用Adam(adaptive moment estimation)优化器,具有较好的收敛速度和稳定性。因为是二分类问题(稳定与不稳定),损失函数使用交叉熵损失(cross-entropy loss)。学习率取0.01,训练部分的批量大小(Batch Size)设为256,因为数据集较大,同时较大的Batch Size可以提高并行性。训练轮数为30轮,使用 Dropout技术(取0.3)防止模型过拟合。
随着智能电网的建设,各级调度中心积累了大量在线安全稳定分析数据,以及广域测量系统(wide area measurement system,WAMS)[23]和相量测量单元(phasor measurement unit,PMU)的广泛普及和应用,这为基于数据驱动方法的电力系统稳定分析和监测提供了可靠的同步数据源。神经网络、决策树、支持向量机、深度学习等数据驱动方法应用到电力系统的暂态稳定评估中,其通过大量离线数据进行训练,建立系统特征与暂态稳定性之间的映射关系。在线应用阶段,故障后根据训练所得的映射关系与 WAMS/PMU 测量数据,快速进行暂态稳定评估,及时为稳定控制算法提供依据。
电力系统暂态稳定性属于大扰动功角稳定问题。其定义为当电力系统发生故障后,若发电机的各功角(δi)能够维持在某一数值附近,则电力系统被认为是暂态稳定的。相反,电力系统暂态不稳定是指在发生较大扰动后,由于发电机之间功率的不平衡,导致机组间的相对功角(δ)非周期性增大,进而引发不稳定现象。由于系统部分发电机加速,部分发电机减速,系统的母线电压也随着功角的改变而改变,所以系统各母线电压幅值及相角可以表征电力系统暂态稳定的特征量。
电力系统暂态稳定指的是系统受到大干扰下的功角稳定。当前较常用的稳定判断依据是扰动发生后转子功角的暂态稳定系数(transient stability index, TSI)[24]
$ \mathrm{TSI}=\frac{180^{\circ}-\left|\Delta \delta_{\max }\right|}{180^{\circ}+\left|\Delta \delta_{\max }\right|} $
式(9)中:Δδmax为任意两台发电机的最大相对功角差。若 Δ δ m a x>180°,则TSI<0,此时可以判定系统失稳,反之,系统稳定。
电力系统暂态稳定问题可以理解为一个二分类问题,即0/1问题,1表示由暂态稳定判据得到的稳定标签,0则表示失稳标签。这时只需选择一系列可以表征电力系统状态的特征集合,然后选择构建好的模型就可以对系统稳定性进行分析。而二分类问题常用的混淆矩阵如表2所示。
在暂态稳定性分析(transient stability assessment,TSA)中,稳定与失稳样本之间的不平衡问题是备受关注的重点之一。失稳样本对模型的准确性以及后续应用可能产生重要影响。除了在生成样本时重点关注失稳状态的发生,还可以通过引入暂态失稳样本的预测准确率(true negative rate, TNR)和 F1-score (F1) 来评估模型性能。F1是衡量模型能力的综合指标。此外,结合整体准确率(accuracy,A)以及暂态稳定样本的预测准确率(true positive rate, TPR)[25],可以更全面地评价模型的性能。它们各指标定义如下。
TNR= T N T N + F P×100%
$R=TNR$
TPR= T P T P + F N×100%
P= T N T N + F N×100%
F1= 2 P R P + R×100%
A= T P + T N T P + F P + T N + F N×100%
主要步骤分为离线训练和在线验证,离线部分通过仿真得到暂态特征的数据,通过暂态稳定判断依据,得到数据所对应的标签值,经过数据处理之后,然后输入所构建的模型Transformer-GAT并行特征融合模型,当模型训练好后,保存模型。之后在PMU在线采集的实时数据作为训练好的模型输入,以此来验证电力系统暂态是否失稳,流程如图5所示。
选用IEEE10机39节点系统为算例进行分析,包含39条母线(节点),系统中有10台发电机,46条输电线路连接母线之间,系统如图6所示。利用Pytorch框架搭建深度学习网络模型,硬件平台为13th Gen Intel© CoreTM i9-13900HX和NVIDIA GeForce RTX 4060 Laptop GPU。
采用的电力系统仿真软件是由西门子PTI公司开发的 PSS/E。该软件能够对电网进行建模、潮流计算以及故障分析,广泛应用于电力系统的规划、运行和分析等领域。PSS/E 不仅支持通过图形用户界面 (graphical user interface, GUI) 设置各种运行场景,还提供与Python的接口,可以自动批量生成暂态稳定样本,从而方便、快速地获得大量仿真数据。
在全接线系统的运行模式下,以5%为步长调整负荷水平,范围为70%~140%。发电机出力随机分配,仿真总时长为3 s。假设在0.1 s时系统发生三相短路故障,并分别在0.2、0.3、0.4、0.5和 0.6 s时切除故障,故障位置随机变化。选取故障发生前、发生时和切除时3个时间节点,仿真得到这3个时刻39个节点的电压标幺值和相角数值。输入节点特征信息的矩阵维度为39×2。最终得到7 000个样本,其中包含3 215个失稳样本。训练集和测试集按7∶3分配。
通过PSSE生成的暂态数据,再经过暂态稳定判断依据TSI对暂态稳定和失稳打上标签,因为数据集较大,所以选取一份失稳IEEE39节点的数据集以及对应的拓扑结构的邻接矩阵。通过对数据进行处理,因为使用的是PyTorch框架,所以要把数据转换成张量形式,从而使得它们可以作为Transformer和GAT并行特征融合模型的输入。
不同于Transformer-GAT并行特征融合,在Transformer和GAT串行方式中,减少了融合方式,如图7所示。
在前向反馈中将暂态时序数据先输入Transformer模型中,再通过线性以及激活函数然后作为GAT模型的输入,同时,电力系统拓扑结构邻接矩阵也一起作为GAT模型输入,最终通过线性空间变化以及softmax输出概率值确定是否暂态失稳。
表3所示,Transformer-GAT并行特征融合模型的准确率(A)在98%以上,同时,F1 、查全率(R)和查准率(P)都在96%以上,相较于Transformer和GAT串行都有了进一步的提升,尤其是在准确率误差上,由原来4%降低到不足2%,误差降低近一半,由于Transformer和GAT串行方法中特征传递过程中的信息可能会丢失,导致最终输出的性能下降,F1 、查全率(R)和查准率(P)都有所下降,说明提出的模型能够同时从暂态数据以及拓扑网络中挖掘到重要特征。
在电力系统中将暂态稳定误判成失稳和将暂态失稳误判成稳定的成本是不一样的,显然后者更严重。为了可视化对比两种模型带来电力系统误判成本,通过对一定批量的稳定和失稳样本所呈现出两种模型的混淆矩阵,来直观对比模型的性能。如图8所示展示了通过使用不同模型方案的混淆矩阵二维图。通过对比可以得到Transformer-GAT并行特征融合提高了对稳定样本和失稳样本的误判,对稳定样本的误判从17份减少到8份,对失稳样本也从13份降低到6份,说明该方案确实可以有效地减少电力系统暂态故障时的误判,降低误判所带来的危害性。
为了验证所构建的Transformer-GAT并行特征融合更加具有有效性和优越性,选取了当前基于数据驱动的电力系统暂态稳定评估的深度学习主流模型,卷积神经网络(convolutional neural network,CNN),循环神经网络(recurrent neural network,RNN)、图卷积神经网络(graph convolutional network,GCN)、Transformer、GAT等模型进行对比。为保证实验一致性,所有模型采用同一份数据集。
通过训练得出不同模型收敛速度如图9所示。
相比于传统的CNN、RNN、GCN、Transformer及GAT模型,提出的Transformer-GAT并行特征融合模型展现出显著的收敛优势。通过图9可知Transformer-GAT模型在第十轮左右趋于收敛,而其他模型如GCN、RNN等一般需要15~20轮后才能达到类似的收敛状态。这一结果表明,Transformer-GAT模型通过并行融合全局依赖和局部节点重要性,大大提升了特征提取的效率,使得模型可以在更少的训练轮次中捕捉到复杂的特征。同时这一快速收敛的特性不仅提高了训练效率,也显著减少了计算资源的消耗,在实际电力系统故障诊断等需要实时响应的场景中具有重要的应用价值。
同时对6种模型进行横向比较,分析得出模型可视化的预测准确率如图10所示以及模型性能指标如表4所示。
通过对准确率的可视化分析,明显看出GAT并行特征融合在性能上具有显著优势。根据表4可知,CNN模型的评估指标低于90%,主要原因是CNN无法有效处理暂态时间序列数据。相比之下,RNN和GCN虽然都能够处理序列特征,但由于GCN还能捕捉拓扑结构信息,模型表现略优于RNN。而基于Transformer和GAT的单一模型,得益于强大的注意力机制,在此基础上评估指标提升了约3%。然而,单一模型在处理特征时可能会丢失部分信息。因此,表4清楚地展示了Transformer-GAT并行特征融合能够显著提升准确率和整体模型评估指标,各项指标Transformer和GAT的单一模型提升幅度接近4%,也验证了所构建模型相比单一模型表现出巨大优势。
为进一步验证所提出模型的鲁棒性,引入暂态电压和相角的噪声数据放入训练集中,以此来评估模型的鲁棒性。实验通过引入100份噪声(随机选取训练集中的100份数据集将标签反转),然后验证模型的鲁棒性。性能指标如图11所示。
所有模型在加入随机噪声后,准确率均有所下降(比未加入噪声之前的Transformer-GAT降低1.7%),尤其是CNN准确率下降近6%,其余几个模型也下降3%~4%,相较于其他模型,Transformer-GAT并行模型性能下降幅度较小,模型表现出了较强的抗噪能力。这也为解决电力系统中存在的暂态虚假数据导致暂态评估准确下降的问题提供了一种方案。以上结果表明,所提出的模型在面对噪声时,表现出了较好的鲁棒性,验证了其在实际应用中的可靠性。
提出了一种基于Transformer-GAT并行特征融合的电力系统暂态稳定性评估方法,针对电力系统暂态仿真数据进行深入挖掘,旨在克服传统方法在捕捉时空特征方面的局限性,经过在典型电力系统中拟合我国暂态工况进行验证,证明了其在暂态稳定性评估中的有效性。主要结论如下。
(1)与Transformer和GAT的串行结构相比,提出的并行特征融合模型在预测准确率上表现更优。并行融合策略有效降低了对稳定样本与失稳样本的误判率,显著减小了因误判引发的潜在风险。
(2)通过与其他5种深度学习模型的比较,本文模型展现出更快的收敛速度,能够满足电力系统实时响应的需求。相较于单独使用Transformer或GAT,提出的融合模型在准确率上提升了3%~4%,对暂态稳定性的识别能力显著增强。
(3)在加入噪声的情况下,模型的并行特征融合方案仍具备较高的鲁棒性。尽管模型的准确率有所下降,但仅下降了1.7%,相比其他模型依然展现出明显的优势。
尽管Transformer-GAT并行特征融合模型在电力系统暂态稳定性评估中取得了显著成果,未来仍有一些潜在的改进方向。首先,可以进一步优化特征融合机制,以更好地平衡时空信息的贡献。其次,除了PMU采集的暂态数据之外,模型的扩展性和适应性可通过引入多模态数据(如气象数据、市场行为数据等)加以增强。此外,在实际应用中,可以通过更多的实时监控数据对模型进行在线更新和调整,从而进一步提升评估的精度与实时性。
  • 国家自然科学基金(62062068)
  • 云南省中青年学术和技术带头人后备人才项目(202305AC160077)
参考文献 引证文献
排序方式:
[1]
中国能源网. 华中电网事故的经验与教训[EB/OL]. ( 2017-08-09) [2025-01-17]. https://www.china5e.com/energy/news-334763-1.html. https://www.china5e.com/energy/news-334763-1.html
China Energy Network. Experience and lessons from the central China power grid accident[EB/OL]. ( 2017-08-09) [2025-01-17]. https://www.china5e.com/energy/news-334763-1.html. https://www.china5e.com/energy/news-334763-1.html
[2]
张智刚, 康重庆. 碳中和目标下构建新型电力系统的挑战与展望[J]. 中国电机工程学报, 2022, 42(8): 2806-2819.
Zhang Zhigang, Kang Chongqing. Challenges and prospects of building a new power system under the carbon neutrality goal[J]. Proceedings of the CSEE, 2022, 42(8): 2806-2819.
[3]
朱冰, 张龙, 司文佳, 等. 跟网型与构网型变流器混合并网的暂态特性分析及故障穿越控制[J]. 科学技术与工程, 2024, 24(28): 12177-12187.
Zhu Bing, Zhang Long, Si Wenjia, et al. Analysis of transient characteristics and fault ride-through control of hybrid grid-tied converters with grid-following and grid-forming[J]. Science Technology and Engineering, 2024, 24(28): 12177-12187.
[4]
高剑, 毕悦, 杨亚兰, 等. 光伏电站机电暂态实用化等值建模方法[J]. 科学技术与工程, 2022, 22(2): 556-562.
Gao Jian, Bi Yue, Yang Yalan, et al. Practical equivalent modeling method of photovoltaic power station in electromechanical transient simulation[J]. Science Technology and Engineering, 2022, 22(2): 556-562.
[5]
戴远航, 陈磊, 张玮灵, 等. 基于多支持向量机综合的电力系统暂态稳定评估[J]. 中国电机工程学报, 2016, 36(5): 1173-1180.
Dai Yuanhang, Chen Lei, Zhang Weiling, et al. Multi-support vector machine power system transient stability assessment based on relief algorithm[J]. Proceedings of the CSEE, 2016, 36(5): 1173-1180.
[6]
刘炼, 王强, 陈浩. 基于改进随机森林的电力系统暂态稳定评估[J]. 科学技术与工程, 2022, 22(11): 4367-4374.
Liu Lian, Wang Qiang, Chen Hao. Transient stability assessment of power systems based on improved random forest[J]. Science Technology and Engineering, 2022, 22(11): 4367-4374.
[7]
曲莹, 韩肖清, 刘新元, 等. 基于多源数据驱动的电力系统暂态稳定性分析方法[J]. 太原理工大学学报, 2024, 55(1): 73-83.
Qu Ying, Han Xiaoqing, Liu Xinyuan, et al. Analysis method for transient stability of power systems based on multi-source data-driven approach[J]. Journal of Taiyuan University of Technology, 2024, 55(1): 73-83.
[8]
李向伟, 刘思言, 高昆仑. 基于差分进化极限学习机的电力系统暂态稳定评估方法[J]. 科学技术与工程, 2020, 20(1): 213-217.
Li Xiangwei, Liu Siyan, Gao Kunlun. Power system transient stability assessment based on the differential evolution extreme learning machine[J]. Science Technology and Engineering, 2020, 20(1): 213-217.
[9]
周芮, 杨燕, 余娟, 等. 适应拓扑变化的数据驱动电力系统暂态主导失稳模式识别方法[J]. 中国电机工程学报, 2025, 45(9): 3436-3448.
Zhou Rui, Yang Yan, Yu Juan, et al. Data-driven method for identifying transient dominant instability patterns in power systems adapted to topological changes[J]. IEEE Transactions on Power Engineering, 2025, 45(9): 3436-3448.
[10]
王强, 陈浩, 刘炼. 基于多层CatBoost的电力系统暂态稳定评估[J]. 科学技术与工程, 2022, 22(4): 1456-1464.
Wang Qiang, Chen Hao, Liu Lian. Research on transient stability assessment of power system based on multi-layer CatBoost[J]. Science Technology and Engineering, 2022, 22(4): 1456-1464.
[11]
彭鑫, 刘俊, 刘嘉诚, 等. 图像化数据驱动的电力系统暂态稳定性在线评估方法[J]. 智慧电力, 2022, 50(11): 17-24.
Peng Xin, Liu Jun, Liu Jiacheng, et al. Data-driven online assessment method for transient stability of power systems based on visualization[J]. Smart Power, 2022, 50(11): 17-24.
[12]
Cheng S, Yu Z, Liu Y, et al. Power system transient stability assessment based on the multiple paralleled convolutional neural network and gated recurrent unit[J]. Protection and Control of Modern Power Systems, 2022, 7(3): 1-16.
[13]
李向伟, 刘思言, 高昆仑. 基于双向长短时记忆网络和卷曲神经网络的电力系统暂态稳定评估[J]. 科学技术与工程, 2020, 20(7): 2733-2739.
Li Xiangwei, Liu Siyan, Gao Kunlun. Power system transient stability assessment based on BiLSTM-ICNN[J]. Science Technology and Engineering, 2020, 20(7): 2733-2739.
[14]
Wang H, Gao F, Chen Q, et al. Instability pattern-guided model updating method for data-driven transient stability assessment[J]. IEEE Transactions on Power Systems, 2025, 40(2): 1214-1227.
[15]
Zhou J, Bu S. TSPM-CNN based power system transient stability assessment using a new data extraction method[C]// 12th IET International Conference on Advances in Power System Control, Operation and Management (APSCOM 2022), Hybrid Conference. Hong Kong: Research Institute for Smart Energy, The Hong Kong Polytechnic University, 2022: 170-175.
[16]
Shao Z, Wang Q, Cao Y, et al. A novel data-driven LSTM-SAF model for power systems transient stability assessment[J]. IEEE Transactions on Industrial Informatics, 2024, 20(7): 9083-9097.
[17]
Umbereen S, Weiss X, Rolander A, et al. Investigating the performance of MLE & CNN for transient stability assessment in power systems[J]. IEEE Access, 2024, 12: 125095-125107.
[18]
房佳姝, 刘崇茹, 苏晨博, 等. 基于自注意力Transformer编码器的多阶段电力系统暂态稳定评估方法[J]. 中国电机工程学报, 2023, 43(15): 5745-5759.
Fang Jiashu, Liu Chongru, Su Chenbo, et al. Multi-stage transient stability assessment method for power systems based on self-attention transformer encoder[J]. IEEE Transactions on Power Engineering, 2023, 43(15): 5745-5759.
[19]
Mahato N K, Dong J, Song C, et al. Electric power system transient stability assessment based on Bi-LSTM attention mechanism[C]// 2021 6th Asia Conference on Power and Electrical Engineering (ACPEE). Chongqing: Engineering Research Center of Energy Electric Power Information Security, 2021: 777-782.
[20]
张亮, 安军, 周毅博. 基于时间卷积和图注意力网络的电力系统暂态稳定评估[J]. 电力系统自动化, 2023, 47(7): 114-122.
Zhang Liang, An Jun, Zhou Yibo. Transient stability assessment of power systems based on temporal convolution and graph attention networks[J]. Automation of Electric Power Systems, 2023, 47(7): 114-122.
[21]
Velikovi P, Cucurull G, Casanova A, et al. Graph attention networks[C]// 6th International Conference on Learning Representations. Cambridge: University of Cambridge, 2018.
[22]
Vaswani A, Shazeer N, Parmar A, et al. Attention is all you need[J]. ArXiv, 2017: 1710.10903.
[23]
段刚, 严亚勤, 谢晓冬, 等. 广域相量测量技术发展现状与展望[J]. 电力系统自动化, 2015, 39(1): 73-80.
Duan Gang, Yan Yaqin, Xie Xiaodong, et al. Current status and prospects of wide-area phasor measurement technology[J]. Automation of Electric Power Systems, 2015, 39(1): 73-80.
[24]
周宇, 肖健梅, 王锡淮. 基于GCN和HGP-SL的电力系统暂态稳定评估[J]. 电气工程学报, 2024, 19(4): 246-254.
Zhou Yu, Xiao Jianmei, Wang Xihuai. Transient stability assessment of power systems based on GCN and HGP-SL[J]. Journal of Electrical Engineering, 2024, 19(4): 246-254.
[25]
王铮澄, 周艳真, 郭庆来, 等. 考虑电力系统拓扑变化的消息传递图神经网络暂态稳定评估[J]. 中国电机工程学报, 2021, 41(7): 2341-2350.
Wang Zhengcheng, Zhou Yanzhen, Guo Qinglai, et al. Transient stability assessment of power systems considering topological changes using message passing graph neural networks[J]. IEEE Transactions on Power Engineering, 2021, 41(7): 2341-2350.
2025年第25卷第21期
PDF下载
244
115
引用本文
BibTeX
文章信息
doi: 10.12404/j.issn.1671-1815.2407675
  • 接收时间:2024-10-15
  • 首发时间:2026-01-13
  • 出版时间:2025-07-28
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-10-15
  • 修回日期:2025-04-22
基金
国家自然科学基金(62062068)
云南省中青年学术和技术带头人后备人才项目(202305AC160077)
作者信息
    1 云南民族大学电气信息工程学院云南省高校电力信息物理融合系统重点实验室, 昆明 650504
    2 云南省无人自主系统重点实验室, 昆明 650504

通讯作者:

* 徐天奇(1978—),男,汉族,云南禄丰人,博士,教授。研究方向:韧性电网、新能源发电并网、电力信息物理系统。E-mail:
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/kxjsygc/CN/10.12404/j.issn.1671-1815.2407675
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