基于深度学习的用电网络绝缘子识别定位技术

打开文本图片集

中图分类号：TP319 文献标志码：A 文章编号：2095-2945（2025）20-0007-05

Abstract：Atpresent，mostoftheidentificationandpositioningofpower lineinsulatorsrelyonmanualwork，whichiscostly andineficient.Thedevelopmentofdeplearningtheoryhasgreatlypromotedtheintellgentidentificationandpositioningof insulators.Thispaper takes theimagerecognitionofinsulatorsinpowersupplynetworksastheresearchobject.Basedonthe deeplearningmethodandcombinedwiththeapplicationcharacteristicsofthepowersupplysafetydetectionandmonitoring devicesystem，theaplicationinintellgentrecognitionofpowersupplysafetydetectionandmoniringimagesisstudiedInthis paper，twoconvolutionalnetworksarecombined.Thenewalgorithmreduces thegenerationofredundantwindowsandusesa moreintellgentslidingmechanismtoimprovepositioningaccuracyThisimprovementmakestheunmannedmonitoringprocesof insulatorsmoreeficient，andtoacertainextentprovidestechnicalsupprtforunmannedreal-timemonitoringof powernetworks.

Keywords： electricity network;contact line; insulator; deep learning;positioning and identification

一般输电线路或铁路接触网中极为重要的组件—一绝缘子，扮演着机械支撑和电气隔离的双重角色。（剩余5478字）