摘要
该文提出了一种基于无源射频识别(radio frequency identification,RFID)振动传感标签及量子粒子群算法(quantum-behaved particle swarm optimization,QPSO)--相关向量机(relevance vector machine,RVM)的变压器绕组故障在线诊断技术。首先设计一种双天线无源RFID振动传感器标签结构,可以稳定工作在无源模式下。针对变压器绕组振动信号包含大量噪声的特点,利用奇异熵对原始信号进行降噪处理,并提出基于QPSO优化的RVM的故障诊断算法。测试结果表明:该文所设计的标签能够可靠地完成变压器绕组振动信号采集以及传输,QPSO-RVM算法能够快速而准确地定位出故障所在,与国内外现有监测技术相比,具有低成本、功耗低,故障定位迅速准确的优点。
An on-line fault diagnosis for transformer windings based on RFID sensor tags and quantum-behaved particle swarm optimization (QPSO)-relevance vector machine (RVM)was proposed in this paper.Firstly,a double antenna radio frequency identification (RFID)sensor tag was designed to detect vibration signals,which could work on passive mode. Considering the large amount of noise components in winding vibration signals,the singular entropy was employed to de-noise the raw signal.The RVM optimized by QPSO was used for fault diagnosis.Experimental results show that the exploited RFID sensor tag can reliably accumulate and transfer the winding vibration signal,and the proposed fault diagnosis approach has merit of accuracy and rapidity.Compared with existing fault diagnosis approaches,the proposed approach has advantages of low cost,low power consumption and can locate the faulty winding quickly and accurately.
作者
邓芳明
温开云
何怡刚
李兵
汪涛
吴翔
DENG Fangming;WEN Kaiyun;HE Yigang;LI Bing;WANG Tao;WU Xiang(School of Electrical and Automation Engineering,East China Jiaotong University,Nanchang 330013,Jiangxi Province,China;School of Electrical Engineering,Wuhan University,Wuhan 430072,Hubei Province,China;School of Electrical Engineering andAutomation,Hefei University of Technology,Hefei 230009,Anhui Province,China)
出处
《中国电机工程学报》
EI
CSCD
北大核心
2018年第24期7183-7193,共11页
Proceedings of the CSEE
关键词
变压器绕组
射频识别
故障诊断
奇异熵
相关向量机
量子粒子群算法
transformer winding
radio frequency identification (RFID)
fault diagnosis
singular entropy
relevance vector machine (RVM)
quantum-behaved particle swarm optimization (QPSO)
