基于混合扫描的碳足迹采集终端可测性设计及融合诊断  

Carbon footprint collection terminal testability design and fusion diagnosis based on hybrid scan

作　　者：赵雪松 尹仕红 谢倩娴 侯婧 林海军[2] 陈寅生 ZHAO Xuesong;YIN Shihong;XIE Qianxian;HOU Jing;LIN Haijun;CHEN Yinsheng(Measurement Management Office,Shenzhen Power Supply Bureau Co.,Ltd.,Shenzhen 518133,Guangdong,China;Harbin University of Science and Technology,Harbin 150080,China)

机构地区：[1]深圳供电局有限公司计量管理所,广东深圳518133 [2]哈尔滨理工大学,哈尔滨150080

出　　处：《电测与仪表》2024年第8期39-46,共8页Electrical Measurement & Instrumentation

基　　金：国家自然科学基金资助项目(61803128)。

摘　　要：在“双碳”战略的背景下,针对国内对碳足迹采集终端及系统的迫切需求,提出了基于电力采集终端及通信系统的解决方案,并利用混合边界扫描技术提出了具体的“虚拟探针”可测性设计方案。还针对基于单一类型故障特征进行非线性“簇”电路故障诊断准确率低的难题,在研究小波包变换、PCA及Volterra核特征提取的基础上,提出了小波包变换与PCA特征层融合,并与基于Volterra核特征的初级诊断结果进行决策层融合的故障诊断方法。实验表明,该方法可以有效提高故障诊断的准确率。Under the background of"carbon peak and neutrality"strategy,in order to solve the domestic urgent demand for the carbon footprint collection terminal and system,a solution based on electricity acquisition terminal and communication system is proposed,and a specific"virtual probe"testability design plan is also put forward by using hybrid boundary scan technology.By focusing on solving the problem of low accuracy in nonlinear"cluster"circuit fault diagnosis based on single type of fault feature,a fault diagnosis method is proposed that involves a fusion of wavelet packet transform and PCA feature layer,based on the study of wavelet packet transform,PCA and Volterra kernel feature extraction.Then,a decision-level fusion is conducted with the primary diagnosis results based on Volterra kernel feature.Experimental results have shown that the proposed method effectively improves the accuracy of fault diagnosis.

关 键 词：碳足迹采集终端 可测性设计 信息融合 故障诊断 VOLTERRA核 

分 类 号：TM93[电气工程—电力电子与电力传动]