Article(id=1325492316114395188, tenantId=1146029695717560320, journalId=1324759967749734599, issueId=1325492191933640907, articleNumber=null, orderNo=null, doi=10.19805/j.cnki.jcspe.2026.250285, pmid=null, cstr=null, oa=null, hot=0, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1747756800000, receivedDateStr=2025-05-21, revisedDate=1752076800000, revisedDateStr=2025-07-10, acceptedDate=null, acceptedDateStr=null, onlineDate=1793951594618, onlineDateStr=2026-11-06, pubDate=null, pubDateStr=null, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1793951594618, onlineIssueDateStr=2026-11-06, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1793951594618, creator=13701087609, updateTime=1793951594618, updator=13701087609, issue=Issue{id=1325492191933640907, tenantId=1146029695717560320, journalId=1324759967749734599, year='2026', volume='46', issue='10', pageStart='1', pageEnd='216', issueExtLink='null', onlineDate='null', pubDate='1791993600000', pubDateStr='2026-10-15', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=1, specialIssue=null, createTime=1793951565011, creator='13701087609', updateTime=1793951662932, updator='13701087609', preIssue=null, nextIssue=null, articleTotal=null, ext={EN=IssueExt(id=1325492603457773727, tenantId=1146029695717560320, journalId=1324759967749734599, issueId=1325492191933640907, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1325492603457773728, tenantId=1146029695717560320, journalId=1324759967749734599, issueId=1325492191933640907, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null, downloadFileDto=null}, startPage=146, endPage=153,216, ext={EN=ArticleExt(id=1325492319922823222, articleId=1325492316114395188, tenantId=1146029695717560320, journalId=1324759967749734599, language=EN, title=Insulator Defect Detection Algorithm Based on Improved YOLOv8-DEW, columnId=null, journalTitle=Journal of Chinese Society of Power Engineering, columnName=null, runingTitle=null, highlight=null, articleAbstract=In transmission corridor inspection, insulator targets are generally small and exhibit significant scale variations, which can easily lead to false detections and missed detections of defects. To address these problems, an improved YOLOv8-dynamic enhanced window (DEW) based insulator defect detection algorithm was proposed. Firstly, a YOLOv8n detection network was constructed. To enhance the adaptive spatial aggregation capability of model for insulator targets with different poses and scales, the third-generation deformable convolution (DCNV3) was introduced. Partial original C2f modules and standard convolution modules in the backbone network were respectively replaced with C2f_DCNV3 and DCNV3_YOLO modules. Secondly, efficient channel attention (ECA) was introduced to enhance the recognition capability of model for target defect features. Finally, wise intersection over union loss (WIOU) was adopted to replace complete intersection over union loss (CIOU), so as to reduce the bounding box regression loss and improve the localization quality of prediction boxes. Experimental results show that the model file size, parameter quantity and computational complexity of the improved model are all slightly reduced, the number of floating-point operations is decreased by 1.3×109, and the detection speed reaches 26 frames/s. For insulator defect detection, the η mAP50 of reaches 98.3%, and the η mAP50:95 reaches 86.6%. Its overall performance is superior to other mainstream object detection algorithms., authors=WANG Daolei1, YANG Tao1, YAN Peng1, LIU Hongda2, ZHU Rui1, authorsList=WANG Daolei, YANG Tao, YAN Peng, LIU Hongda, ZHU Rui, authorCompany=1 College of Computer Science and Technology, Shanghai University of Electric Power, Shanghai 201306, China;
2 State Grid Ningbo Power Supply Company, Ningbo 315000, Zhejiang Province, China, correspAuthors=null, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=null, magXml=null, pdfUrl=null, pdf=null, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=null, mapNumber=null, fund=null), CN=ArticleExt(id=1325492319495004213, articleId=1325492316114395188, tenantId=1146029695717560320, journalId=1324759967749734599, language=CN, title=基于改进YOLOv8-DEW的绝缘子缺陷检测算法, columnId=1325492212859023620, journalTitle=动力工程学报, columnName=数字化与智能化, runingTitle=null, highlight=null, articleAbstract=在输电通道巡检中,绝缘子目标尺寸较小、尺度差异明显,易造成缺陷误检和漏检,提出一种基于改进YOLOv8-动态增强窗口(DEW)的绝缘子缺陷检测算法。首先,搭建YOLOv8n检测网络,为增强模型对不同姿态和尺度绝缘子目标的自适应空间聚合能力,引入第三代可变形卷积(DCNV3),将主干网络中原有的部分C2f模块和标准卷积模块分别替换为C2f_DCNV3和DCNV3_YOLO模块;其次,引入高效通道注意力(ECA),增强模型对目标缺陷特征的识别能力;最后,采用智慧交并比损失(WIOU)替代完全交并比损失(CIOU),减小边界框回归损失,提高预测框定位质量。实验结果表明:改进模型的模型文件大小、参数量和计算量均略有减小,浮点运算次数减少1.3×10⁹次,检测速度达到26帧/s;绝缘子缺陷检测的指标η mAP₅₀达到98.3%,指标η mAP₅₀:₉₅达到86.6%,其整体性能优于其他主流目标检测算法。, authors=王道累1, 杨涛1, 闫鹏1, 刘宏达2, 朱瑞1, authorsList=王道累, 杨涛, 闫鹏, 刘宏达, 朱瑞, authorCompany=1 上海电力大学 计算机科学与技术学院, 上海 201306;
2 国网浙江省电力有限公司宁波供电公司, 浙江宁波 315000, correspAuthors=null, authorNote=王道累: 王道累(1981—),男,上海人,教授,博士,研究方向为动力设备故障诊断、智能化方法研究。朱瑞(通信作者),女,教授,博士,E-mail: zhurui@shiep.edu.cn。, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=null, magXml=null, pdfUrl=null, pdf=fLfOl8DwAZKa4UOh+ZQvRw==, pdfFileSize=4298421, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=null, mapNumber=null, fund=国家自然科学基金资助项目 (12172210,61502297))}, authors=null, keywords=[Keyword(id=1325492321655070776, tenantId=1146029695717560320, journalId=1324759967749734599, articleId=1325492316114395188, language=CN, orderNo=1, keyword=缺陷检测), Keyword(id=1325492321743151162, tenantId=1146029695717560320, journalId=1324759967749734599, articleId=1325492316114395188, language=CN, orderNo=2, keyword=绝缘子), Keyword(id=1325492322053529659, tenantId=1146029695717560320, journalId=1324759967749734599, articleId=1325492316114395188, language=CN, orderNo=3, keyword=YOLOv8n), Keyword(id=1325492322414239804, tenantId=1146029695717560320, journalId=1324759967749734599, articleId=1325492316114395188, language=CN, orderNo=4, keyword=注意力), Keyword(id=1325492322493931581, tenantId=1146029695717560320, journalId=1324759967749734599, articleId=1325492316114395188, language=CN, orderNo=5, keyword=可变形卷积), Keyword(id=1325492323248906302, tenantId=1146029695717560320, journalId=1324759967749734599, articleId=1325492316114395188, language=EN, orderNo=1, keyword=fault detection), Keyword(id=1325492323336986687, tenantId=1146029695717560320, journalId=1324759967749734599, articleId=1325492316114395188, language=EN, orderNo=2, keyword=insulator), Keyword(id=1325492323756417088, tenantId=1146029695717560320, journalId=1324759967749734599, articleId=1325492316114395188, language=EN, orderNo=3, keyword=YOLOv8n), Keyword(id=1325492324133904449, tenantId=1146029695717560320, journalId=1324759967749734599, articleId=1325492316114395188, language=EN, orderNo=4, keyword=attention), Keyword(id=1325492324515586114, tenantId=1146029695717560320, journalId=1324759967749734599, articleId=1325492316114395188, language=EN, orderNo=5, keyword=deformable convolution)], refs=null, funds=null, companyList=null, figs=null, attaches=null, journal=Journal(id=1324759778016198853, delFlag=0, nameCn=动力工程学报, nameEn=Journal of Chinese Society of Power Engineering, nameHistory1=null, nameHistory2=null, issn=1674-7607, eissn=null, cn=31-2041, coden=null, periodic=0, language=CN, oaType=null, ccby=null, superviseOffice=null, ownerOffice=null, pubOffice=null, editorOffice=null, officeType=null, aims=null, clcCode=null, officeProv=null, officeCity=null, officeAddr=null, officeZip=null, officeEmail=null, officePhone=null, editDirector=null, officeDirector=null, officeDirectorPhone=null, officeStaffNum=null, officeEmpNum=null, coverPicUrl=Z8gmCdxHtRC6BlwMqQPuLw==, journalPrice=null, startedYear=null, abbrevIsoEn=Journal of Chinese Society of Power Engineering, journalRemark=null, publicationField=null, createdTime=1793776943927, updatedTime=1793850768655, createdBy=18614031015, updatedBy=13701087609, firstLetterCn=D, firstLetterEn=D, subjectCode=Engineering, subjectName=null, subjectCodeEn=Engineering, subjectNameEn=null, picCn=Z8gmCdxHtRC6BlwMqQPuLw==, picEn=GHlzlzY6AAwEqnHFaQScZg==, jcr=null, cjcr=null, exts=[JournalExt(id=1325069421433201525, language=CN, name=动力工程学报, 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=1793850768669, updatedTime=1793850768669, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://jcspe1.speri.com.cn/Journalx_dlgcxb/authorLogOn.action, submissionEditorUrl=https://jcspe1.speri.com.cn/Journalx_dlgcxb/editorLogOn.action, submissionReviewUrl=https://jcspe1.speri.com.cn/Journalx_dlgcxb/expertLogOn.action, submissionCeEditorUrl=https://jcspe1.speri.com.cn/Journalx_dlgcxb/editorInChiefLogOn.action, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1325069421521281910, language=EN, name=Journal of Chinese Society of Power Engineering, 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=1793850768690, updatedTime=1793850768690, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://jcspe1.speri.com.cn/Journalx_dlgcxb/authorLogOn.action, submissionEditorUrl=https://jcspe1.speri.com.cn/Journalx_dlgcxb/editorLogOn.action, submissionReviewUrl=https://jcspe1.speri.com.cn/Journalx_dlgcxb/expertLogOn.action, submissionCeEditorUrl=https://jcspe1.speri.com.cn/Journalx_dlgcxb/editorInChiefLogOn.action, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1324759967749734599, websiteList=[Website(id=1325070300190224423, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1324759967749734599, 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/dlgcxb/CN, language=CN, createTime=1793850978181, createBy=18614031015, updateTime=1793851174168, updateBy=18614031015, name=动力工程学报-中文, tplId=1146099689490845704, title=动力工程学报, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1325071272366977204, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300190224423, code=articleTextType, value=kx, createTime=1793851209966, updateTime=1793851209966, creator=18614031015, updator=18614031015), WebsiteProps(id=1325071272341811377, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300190224423, code=banner, value=null, createTime=1793851209960, updateTime=1793851209960, creator=18614031015, updator=18614031015), WebsiteProps(id=1325071272392143031, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300190224423, code=grayFlag, value=0, createTime=1793851209972, updateTime=1793851209972, creator=18614031015, updator=18614031015), WebsiteProps(id=1325071272329228464, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300190224423, code=logo, value=https://castjournals.cast.org.cn/joweb/dlgcxb/CN/file/pic?fileId=YkysOzwd4AobFulcCLHFQg==, createTime=1793851209957, updateTime=1793851209957, creator=18614031015, updator=18614031015), WebsiteProps(id=1325071272408920249, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300190224423, code=minRunFlag, value=0, createTime=1793851209976, updateTime=1793851209976, creator=18614031015, updator=18614031015), WebsiteProps(id=1325071272358588595, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300190224423, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/dlgcxb/CN/file/pic, createTime=1793851209964, updateTime=1793851209964, creator=18614031015, updator=18614031015), WebsiteProps(id=1325071272400531640, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300190224423, code=silenceFlag, value=0, createTime=1793851209974, updateTime=1793851209974, creator=18614031015, updator=18614031015), WebsiteProps(id=1325071272350199986, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300190224423, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1793851209962, updateTime=1793851209962, creator=18614031015, updator=18614031015), WebsiteProps(id=1325071272375365813, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300190224423, code=themeColor, value=null, createTime=1793851209968, updateTime=1793851209968, creator=18614031015, updator=18614031015), WebsiteProps(id=1325071272383754422, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300190224423, code=themeStyle, value=null, createTime=1793851209970, updateTime=1793851209970, creator=18614031015, updator=18614031015)]), Website(id=1325070300584489003, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1324759967749734599, 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/dlgcxb/EN, language=EN, createTime=1793850978275, createBy=18614031015, updateTime=1793851170033, updateBy=18614031015, name=动力工程学报-英文, tplId=1146101810881728533, title=Journal of Chinese Society of Power Engineering, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1325071240246997159, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300584489003, code=articleTextType, value=kx, createTime=1793851202308, updateTime=1793851202308, creator=18614031015, updator=18614031015), WebsiteProps(id=1325071239789818020, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300584489003, code=banner, value=null, createTime=1793851202199, updateTime=1793851202199, creator=18614031015, updator=18614031015), WebsiteProps(id=1325071240523821226, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300584489003, code=grayFlag, value=0, createTime=1793851202374, updateTime=1793851202374, creator=18614031015, updator=18614031015), WebsiteProps(id=1325071239773040803, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300584489003, code=logo, value=https://castjournals.cast.org.cn/joweb/dlgcxb/EN/file/pic?fileId=YkysOzwd4AobFulcCLHFQg==, createTime=1793851202195, updateTime=1793851202195, creator=18614031015, updator=18614031015), WebsiteProps(id=1325071240691593388, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300584489003, code=minRunFlag, value=0, createTime=1793851202414, updateTime=1793851202414, creator=18614031015, updator=18614031015), WebsiteProps(id=1325071240033087654, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300584489003, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/dlgcxb/EN/file/pic, createTime=1793851202257, updateTime=1793851202257, creator=18614031015, updator=18614031015), WebsiteProps(id=1325071240649650347, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300584489003, code=silenceFlag, value=0, createTime=1793851202404, updateTime=1793851202404, creator=18614031015, updator=18614031015), WebsiteProps(id=1325071239945007269, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300584489003, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1793851202236, updateTime=1793851202236, creator=18614031015, updator=18614031015), WebsiteProps(id=1325071240385409192, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300584489003, code=themeColor, value=null, createTime=1793851202341, updateTime=1793851202341, creator=18614031015, updator=18614031015), WebsiteProps(id=1325071240481878185, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300584489003, code=themeStyle, value=null, createTime=1793851202364, updateTime=1793851202364, creator=18614031015, updator=18614031015)])], journalTitle=动力工程学报, weixinUrl=null, journalUrl=https://jcspe.speri.com.cn/, iacademicId=null, status=1, seqNo=null, journalTitleEn=Journal of Chinese Society of Power Engineering, journalPhotoCn=Z8gmCdxHtRC6BlwMqQPuLw==, journalPhotoEn=GHlzlzY6AAwEqnHFaQScZg==, journalFirstLetter=D, journalRecommend=null, journalNew=null, journalCollection=null, jcrJf=null, cjcrJf=null, jcrJfStr=null, cjcrJfStr=null, submissionFirstDecision=null, sciSubjectClassification=null, casSubjectClassification=null, citeScore=null, totalCitationFrequency=null, icpCode=null, psCode=null, advertisingLicenseCode=null, copyrightInformation=null, country=null, option=, provinceCode=null, provinceName=null, collectFlag=false, interPubPlatform=, interPubPlatformUrl=null), detailUrlCn=https://castjournals.cast.org.cn/joweb/dlgcxb/CN/10.19805/j.cnki.jcspe.2026.250285, detailUrlEn=https://castjournals.cast.org.cn/joweb/dlgcxb/EN/10.19805/j.cnki.jcspe.2026.250285, pdfUrlCn=https://castjournals.cast.org.cn/joweb/dlgcxb/CN/PDF/10.19805/j.cnki.jcspe.2026.250285, pdfUrlEn=https://castjournals.cast.org.cn/joweb/dlgcxb/EN/PDF/10.19805/j.cnki.jcspe.2026.250285, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, previewStatus=0, delFlag=0, hasFullText=0, orderTime=1793951594618, fullTextJson=null, articleText=null, reference=刘云鹏, 李泳霖, 裴少通, 等. 基于紫外光辐射照度特征的污秽瓷绝缘子绝缘状态评估方法[J]. 高电压技术, 2023, 49(4): 1622-1631. LIU Yunpeng, LI Yonglin, PEI Shaotong, et al. Evaluation of the insulation state of contaminated porcelain insulators based on character-istics of ultraviolet irradiance[J]. High Voltage Engineering, 2023, 49(4): 1622-1631.
尹向雷, 解永芳, 屈少鹏, 等. 基于动态蛇形卷积和非跨步卷积的绝缘子缺陷检测[J]. 电力系统保护与控制, 2024, 52(20): 177-187. YIN Xianglei, XIE Yongfang, QU Shaopeng, et al. Insulator defect detection based on dynamic snake convolution and space-to-depth convolution[J]. Power System Protection and Control, 2024, 52(20): 177-187.
魏贤哲, 卢武, 赵文彬, 等. 基于改进Mask R-CNN的输电线路防外破目标检测方法研究[J]. 电力系统保护与控制, 2021, 49(23): 155-162. WEI Xianzhe, LU Wu, ZHAO Wenbin, et al. Target detection method for external damage of a transmission line based on an improved Mask R-CNN algorithm[J]. Power System Protection and Control, 2021, 49(23): 155-162.
张伟业, 缪维跑, 苏欢欢, 等. 基于不平衡样本数据的轴承故障诊断模型与方法研究[J]. 动力工程学报, 2025, 45(6): 904-912. ZHANG Weiye, MIAO Weipao, SU Huanhuan, et al. Research on bearing fault diagnosis model and method based on unbalanced sample data[J]. Journal of Chinese Society of Power Engineering, 2025, 45(6): 904-912.
章力, 邓艾东, 王敏, 等. 基于通道注意力机制与多尺度减法轻量化网络的滚动轴承故障诊断[J]. 动力工程学报, 2025, 45(4): 571-581. ZHANG Li, DENG Aidong, WANG Min, et al. Fault diagnosis of rolling bearing based on channel attention mechanism and multi-scale subtraction lightweight neural network[J]. Journal of Chinese Society of Power Engineering, 2025, 45(4): 571-581.
TAO Chongben, CAO Chunli, CHENG Hanjing, et al. An efficient 3D object detection method based onfast guided anchor stereo RCNN[J]. Advanced Engineering Informatics, 2023, 57: 102069.
HE Zhige, HE Yuanqing. AS-Faster-RCNN: an improved object detection algorithm for airport scene based on faster R-CNN[J]. IEEE Access, 2025, 13: 36050-36064.
TAN Ling, WU Hui, XU Zifeng, et al. Multi-object garbage image detection algorithm based on SP-SSD[J]. Expert Systems with Applications, 2025, 263: 125773.
LIU Bao, JIANG Wenqiang. LA-YOLO: bidirectional adaptive feature fusion approach for small object detection of insulator self-explosion defects[J]. IEEE Transactions on Power Delivery, 2024, 39(6): 3387-3397.
REDMON J, FARHADI A. YOLOv3: an incremental improvement[EB/OL]. (2018-04-08)[2026-07-14]. https://arxiv.org/abs/1804.02767.
康兆楠. 基于YOLOv2的空间场景目标检测算法研究与实现[D]. 北京: 北京交通大学, 2022. KANG Zhaonan. Research and implementation of spatial scene target detection algorithm based on YOLOv2[D]. Beijing: Beijing Jiaotong University, 2022.
陈科羽, 徐梁刚, 陈凤翔, 等. 基于YOLOv3的航拍输电通道航拍绝缘子检测方法研究[J]. 科技创新与应用, 2023, 13(11): 34-37.CHEN Keyu, XU Lianggang, CHEN Fengxiang, et al. Research on a YOLOv3-based method for detecting insulators in aerial images of transmission corridors[J]. Technology Innovation and Application, 2023, 13(11): 34-37.
李运堂, 詹叶君, 王鹏峰, 等. 基于改进YOLOV4网络的绝缘子缺陷检测[J]. 传感技术学报, 2023, 36(8): 1250-1260. LI Yuntang, ZHAN Yejun, WANG Pengfeng, et al. Inspection method for insulator defect based on improved YOLOV4 network[J]. Chinese Journal of Sensors and Actuators, 2023, 36(8): 1250-1260.
王道累, 张正刚, 张世恒, 等. 基于密集连接网络的航拍绝缘子旋转目标精准定位方法[J]. 电力系统保护与控制, 2024, 52(1): 35-43. WANG Daolei, ZHANG Zhenggang, ZHANG Shiheng, et al. Accurate positioning method of insulator rotating target in aerial photography based on dense connection network[J]. Power System Protection andControl, 2024, 52(1): 35-43.
张剑锐, 魏霞, 张林鍹, 等. 改进YOLO v7的绝缘子检测与定位[J]. 计算机工程与应用, 2024, 60(4): 183-191. ZHANG Jianrui, WEI Xia, ZHANG Linxuan, et al. Improving detectionand positioning of insulators in YOLO v7[J]. Computer Engineering and Applications, 2024, 60(4): 183-191.
VARGHESE R, SAMBATH M. YOLOv8: a novel object detection algorithm with enhanced performance and robustness[C]//Proceeding of the 2024 International Conference on Advances in Data Engineering and Intelligent Computing Systems (ADICS). Chennai, India: IEEE, 2024: 1-6.
王长刚, 王先伟, 曹宇, 等. 基于改进图注意力网络的电力系统脆弱性关键环节辨识[J]. 电力系统保护与控制, 2024, 52(15): 36-45. WANG Changgang, WANG Xianwei, CAO Yu, et al. Critical link identification of power system vulnerability based on modified graph attention network[J]. Power System Protection and Control, 2024, 52(15): 36-45.
陈思雨, 付章杰. 融合高效注意力的多尺度输电线路部件检测[J]. 计算机工程与应用, 2024, 60(1): 327-336. CHEN Siyu, FU Zhangjie. Multi-scale transmission line component detection incorporating efficient attention[J]. Computer Engineering and Applications, 2024, 60(1): 327-336.
杜耀文, 谢静, 刘志坚, 等. 基于深度学习的高压隔离开关分合状态检测算法研究[J]. 电力系统保护与控制, 2023, 51(19): 114-123. DU Yaowen, XIE Jing, LIU Zhijian, et al. A detection algorithm for opening and closing states of high-voltage isolation switches based on deep learning[J]. Power System Protection and Control, 2023, 51(19): 114-123.
张秀再, 沈涛, 许岱. 基于改进YOLOv8算法的遥感图像目标检测[J]. 激光与光电子学进展, 2024, 61(10): 1028001.ZHANG Xiuzai, SHEN Tao, XU Dai. Remote-sensing image object detection based on improved YOLOv8 algorithm[J]. Laser & Optoelectronics Progress, 2024, 61(10): 1028001.
龙玉江, 卫薇, 舒彧, 等. 基于自适应关键点的破损旋转绝缘子检测方法[J]. 计算机工程, 2023, 49(9): 272-278. LONG Yujiang, WEI Wei, SHU Yu, et al. Detection method for damaged rotating insulator based on adaptive key points[J]. Computer Engineering, 2023, 49(9): 272-278.
TAO Xian, ZHANG Dapeng, WANG Zihao, et al. Detection of power line insulator defects using aerial images analyzed with convolutional neural networks[J]. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2020, 50(4): 1486-1498.)
收藏切换
基于改进YOLOv8-DEW的绝缘子缺陷检测算法
动力工程学报 | 数字化与智能化 2026,46(10): 146-153,216
收起
收藏切换
动力工程学报 |数字化与智能化 2026 , 46 (10) : 146 -153,216
基于改进YOLOv8-DEW的绝缘子缺陷检测算法
全屏
王道累1, 杨涛1, 闫鹏1, 刘宏达2, 朱瑞1
作者信息
    1 上海电力大学 计算机科学与技术学院, 上海 201306;
    2 国网浙江省电力有限公司宁波供电公司, 浙江宁波 315000
作者简介:
王道累: 王道累(1981—),男,上海人,教授,博士,研究方向为动力设备故障诊断、智能化方法研究。朱瑞(通信作者),女,教授,博士,E-mail: zhurui@shiep.edu.cn。
Insulator Defect Detection Algorithm Based on Improved YOLOv8-DEW
  • WANG Daolei1, YANG Tao1, YAN Peng1, LIU Hongda2, ZHU Rui1
  • Affiliations
      1 College of Computer Science and Technology, Shanghai University of Electric Power, Shanghai 201306, China;
      2 State Grid Ningbo Power Supply Company, Ningbo 315000, Zhejiang Province, China
    doi: 10.19805/j.cnki.jcspe.2026.250285
    文章导航
    收藏切换
    在输电通道巡检中,绝缘子目标尺寸较小、尺度差异明显,易造成缺陷误检和漏检,提出一种基于改进YOLOv8-动态增强窗口(DEW)的绝缘子缺陷检测算法。首先,搭建YOLOv8n检测网络,为增强模型对不同姿态和尺度绝缘子目标的自适应空间聚合能力,引入第三代可变形卷积(DCNV3),将主干网络中原有的部分C2f模块和标准卷积模块分别替换为C2f_DCNV3和DCNV3_YOLO模块;其次,引入高效通道注意力(ECA),增强模型对目标缺陷特征的识别能力;最后,采用智慧交并比损失(WIOU)替代完全交并比损失(CIOU),减小边界框回归损失,提高预测框定位质量。实验结果表明:改进模型的模型文件大小、参数量和计算量均略有减小,浮点运算次数减少1.3×10⁹次,检测速度达到26帧/s;绝缘子缺陷检测的指标η mAP₅₀达到98.3%,指标η mAP₅₀:₉₅达到86.6%,其整体性能优于其他主流目标检测算法。
    缺陷检测  /  绝缘子  /  YOLOv8n  /  注意力  /  可变形卷积
    In transmission corridor inspection, insulator targets are generally small and exhibit significant scale variations, which can easily lead to false detections and missed detections of defects. To address these problems, an improved YOLOv8-dynamic enhanced window (DEW) based insulator defect detection algorithm was proposed. Firstly, a YOLOv8n detection network was constructed. To enhance the adaptive spatial aggregation capability of model for insulator targets with different poses and scales, the third-generation deformable convolution (DCNV3) was introduced. Partial original C2f modules and standard convolution modules in the backbone network were respectively replaced with C2f_DCNV3 and DCNV3_YOLO modules. Secondly, efficient channel attention (ECA) was introduced to enhance the recognition capability of model for target defect features. Finally, wise intersection over union loss (WIOU) was adopted to replace complete intersection over union loss (CIOU), so as to reduce the bounding box regression loss and improve the localization quality of prediction boxes. Experimental results show that the model file size, parameter quantity and computational complexity of the improved model are all slightly reduced, the number of floating-point operations is decreased by 1.3×109, and the detection speed reaches 26 frames/s. For insulator defect detection, the η mAP50 of reaches 98.3%, and the η mAP50:95 reaches 86.6%. Its overall performance is superior to other mainstream object detection algorithms.
    fault detection  /  insulator  /  YOLOv8n  /  attention  /  deformable convolution
    王道累, 杨涛, 闫鹏, 刘宏达, 朱瑞. 基于改进YOLOv8-DEW的绝缘子缺陷检测算法. 动力工程学报, 2026 , 46 (10) : 146 -153,216 . DOI: 10.19805/j.cnki.jcspe.2026.250285
    WANG Daolei, YANG Tao, YAN Peng, LIU Hongda, ZHU Rui. Insulator Defect Detection Algorithm Based on Improved YOLOv8-DEW[J]. Journal of Chinese Society of Power Engineering, 2026 , 46 (10) : 146 -153,216 . DOI: 10.19805/j.cnki.jcspe.2026.250285

      国家自然科学基金资助项目 (12172210,61502297)

    参考文献 引证文献
    排序方式:
    刘云鹏, 李泳霖, 裴少通, 等. 基于紫外光辐射照度特征的污秽瓷绝缘子绝缘状态评估方法[J]. 高电压技术, 2023, 49(4): 1622-1631. LIU Yunpeng, LI Yonglin, PEI Shaotong, et al. Evaluation of the insulation state of contaminated porcelain insulators based on character-istics of ultraviolet irradiance[J]. High Voltage Engineering, 2023, 49(4): 1622-1631.
    尹向雷, 解永芳, 屈少鹏, 等. 基于动态蛇形卷积和非跨步卷积的绝缘子缺陷检测[J]. 电力系统保护与控制, 2024, 52(20): 177-187. YIN Xianglei, XIE Yongfang, QU Shaopeng, et al. Insulator defect detection based on dynamic snake convolution and space-to-depth convolution[J]. Power System Protection and Control, 2024, 52(20): 177-187.
    魏贤哲, 卢武, 赵文彬, 等. 基于改进Mask R-CNN的输电线路防外破目标检测方法研究[J]. 电力系统保护与控制, 2021, 49(23): 155-162. WEI Xianzhe, LU Wu, ZHAO Wenbin, et al. Target detection method for external damage of a transmission line based on an improved Mask R-CNN algorithm[J]. Power System Protection and Control, 2021, 49(23): 155-162.
    张伟业, 缪维跑, 苏欢欢, 等. 基于不平衡样本数据的轴承故障诊断模型与方法研究[J]. 动力工程学报, 2025, 45(6): 904-912. ZHANG Weiye, MIAO Weipao, SU Huanhuan, et al. Research on bearing fault diagnosis model and method based on unbalanced sample data[J]. Journal of Chinese Society of Power Engineering, 2025, 45(6): 904-912.
    章力, 邓艾东, 王敏, 等. 基于通道注意力机制与多尺度减法轻量化网络的滚动轴承故障诊断[J]. 动力工程学报, 2025, 45(4): 571-581. ZHANG Li, DENG Aidong, WANG Min, et al. Fault diagnosis of rolling bearing based on channel attention mechanism and multi-scale subtraction lightweight neural network[J]. Journal of Chinese Society of Power Engineering, 2025, 45(4): 571-581.
    TAO Chongben, CAO Chunli, CHENG Hanjing, et al. An efficient 3D object detection method based onfast guided anchor stereo RCNN[J]. Advanced Engineering Informatics, 2023, 57: 102069.
    HE Zhige, HE Yuanqing. AS-Faster-RCNN: an improved object detection algorithm for airport scene based on faster R-CNN[J]. IEEE Access, 2025, 13: 36050-36064.
    TAN Ling, WU Hui, XU Zifeng, et al. Multi-object garbage image detection algorithm based on SP-SSD[J]. Expert Systems with Applications, 2025, 263: 125773.
    LIU Bao, JIANG Wenqiang. LA-YOLO: bidirectional adaptive feature fusion approach for small object detection of insulator self-explosion defects[J]. IEEE Transactions on Power Delivery, 2024, 39(6): 3387-3397.
    REDMON J, FARHADI A. YOLOv3: an incremental improvement[EB/OL]. (2018-04-08)[2026-07-14]. https://arxiv.org/abs/1804.02767.
    康兆楠. 基于YOLOv2的空间场景目标检测算法研究与实现[D]. 北京: 北京交通大学, 2022. KANG Zhaonan. Research and implementation of spatial scene target detection algorithm based on YOLOv2[D]. Beijing: Beijing Jiaotong University, 2022.
    陈科羽, 徐梁刚, 陈凤翔, 等. 基于YOLOv3的航拍输电通道航拍绝缘子检测方法研究[J]. 科技创新与应用, 2023, 13(11): 34-37.CHEN Keyu, XU Lianggang, CHEN Fengxiang, et al. Research on a YOLOv3-based method for detecting insulators in aerial images of transmission corridors[J]. Technology Innovation and Application, 2023, 13(11): 34-37.
    李运堂, 詹叶君, 王鹏峰, 等. 基于改进YOLOV4网络的绝缘子缺陷检测[J]. 传感技术学报, 2023, 36(8): 1250-1260. LI Yuntang, ZHAN Yejun, WANG Pengfeng, et al. Inspection method for insulator defect based on improved YOLOV4 network[J]. Chinese Journal of Sensors and Actuators, 2023, 36(8): 1250-1260.
    王道累, 张正刚, 张世恒, 等. 基于密集连接网络的航拍绝缘子旋转目标精准定位方法[J]. 电力系统保护与控制, 2024, 52(1): 35-43. WANG Daolei, ZHANG Zhenggang, ZHANG Shiheng, et al. Accurate positioning method of insulator rotating target in aerial photography based on dense connection network[J]. Power System Protection andControl, 2024, 52(1): 35-43.
    张剑锐, 魏霞, 张林鍹, 等. 改进YOLO v7的绝缘子检测与定位[J]. 计算机工程与应用, 2024, 60(4): 183-191. ZHANG Jianrui, WEI Xia, ZHANG Linxuan, et al. Improving detectionand positioning of insulators in YOLO v7[J]. Computer Engineering and Applications, 2024, 60(4): 183-191.
    VARGHESE R, SAMBATH M. YOLOv8: a novel object detection algorithm with enhanced performance and robustness[C]//Proceeding of the 2024 International Conference on Advances in Data Engineering and Intelligent Computing Systems (ADICS). Chennai, India: IEEE, 2024: 1-6.
    王长刚, 王先伟, 曹宇, 等. 基于改进图注意力网络的电力系统脆弱性关键环节辨识[J]. 电力系统保护与控制, 2024, 52(15): 36-45. WANG Changgang, WANG Xianwei, CAO Yu, et al. Critical link identification of power system vulnerability based on modified graph attention network[J]. Power System Protection and Control, 2024, 52(15): 36-45.
    陈思雨, 付章杰. 融合高效注意力的多尺度输电线路部件检测[J]. 计算机工程与应用, 2024, 60(1): 327-336. CHEN Siyu, FU Zhangjie. Multi-scale transmission line component detection incorporating efficient attention[J]. Computer Engineering and Applications, 2024, 60(1): 327-336.
    杜耀文, 谢静, 刘志坚, 等. 基于深度学习的高压隔离开关分合状态检测算法研究[J]. 电力系统保护与控制, 2023, 51(19): 114-123. DU Yaowen, XIE Jing, LIU Zhijian, et al. A detection algorithm for opening and closing states of high-voltage isolation switches based on deep learning[J]. Power System Protection and Control, 2023, 51(19): 114-123.
    张秀再, 沈涛, 许岱. 基于改进YOLOv8算法的遥感图像目标检测[J]. 激光与光电子学进展, 2024, 61(10): 1028001.ZHANG Xiuzai, SHEN Tao, XU Dai. Remote-sensing image object detection based on improved YOLOv8 algorithm[J]. Laser & Optoelectronics Progress, 2024, 61(10): 1028001.
    龙玉江, 卫薇, 舒彧, 等. 基于自适应关键点的破损旋转绝缘子检测方法[J]. 计算机工程, 2023, 49(9): 272-278. LONG Yujiang, WEI Wei, SHU Yu, et al. Detection method for damaged rotating insulator based on adaptive key points[J]. Computer Engineering, 2023, 49(9): 272-278.
    TAO Xian, ZHANG Dapeng, WANG Zihao, et al. Detection of power line insulator defects using aerial images analyzed with convolutional neural networks[J]. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2020, 50(4): 1486-1498.
    2026年第46卷第10期
    112
    18
    0
    引用本文
    BibTeX
    文章信息
    doi: 10.19805/j.cnki.jcspe.2026.250285
    • 接收时间:2025-05-21
    • 首发时间:2026-11-06
    补充材料
    相关文章
    文章信息
    作者
    出版历史
    • 收稿日期:2025-05-21
    • 修回日期:2025-07-10
    基金
    作者信息
    参考文献
    分享链接
    https://castjournals.cast.org.cn/joweb/dlgcxb/CN/10.19805/j.cnki.jcspe.2026.250285
    分享至
    全文二维码

    扫描看全文

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