特高压换流变真空有载分接开关振动及电机电流信号自适应分离方法  

Adaptive separation method of vibration and motor current signal of vacuum on-load tap-changer for converter transformer

作　　者：何良 丁登伟 李献伟[2] 袁明虎 张紫薇 刘卫东[3] HE Liang;DING Dengwei;LI Xianwei;YUAN Minghu;ZHANG Ziwei;LIU Weidong(Sichuan Energy Internet Research Institute Tsinghua University,Chengdu 610213,China;Shanghai Humming Electric Equipment Manufacturing Co.,Ltd.,Shanghai 200333,China;Department of Electrical Engineering,Tsinghua University,Beijing 100084,China)

机构地区：[1]清华四川能源互联网研究院,四川成都610213 [2]上海华明电力设备制造有限公司,上海200333 [3]清华大学电机工程与应用电子技术系,北京100084

出　　处：《电机与控制学报》2023年第8期73-79,共7页Electric Machines and Control

基　　金：国家电网公司总部科技项目(5500-202016076A-0-0-00)。

摘　　要：有载分接开关(OLTC)是特高压(UHV)换流变压器的核心部件,基于驱动电机电流和振动信号的OLTC机械状态在线监测技术是近年来研究热点。为了将振动信号和驱动电机电流信号片段从OLTC长时在线监测数据中自动提取出来,本文提出基于端点检测的信号自适应分离方法。首先,介绍端点检测基本原理,提出振动信号及电机电流信号分离流程;然后,搭建OLTC在线监测系统,并在特高压换流站实际部署;最后,分析在线监测信号时域、频域特征,采用差异化的端点检测方法分别分离出OLTC振动信号和电机电流信号,并结合信号起止点划分驱动电机的3个工作阶段。结果表明,该方法能实现OLTC振动及电流电机信号片段自适应分离,定位误差小于5 ms,为OLTC在线监测数据分析和故障诊断提供基础。On-load tap-changer(OLTC)is the core component of ultra high voltage(UHV)converter transformer,and the online monitoring technology based on driving motor current and vibration signal is a research hotspot for OLTC mechanical state in recent years.In order to automatically extract the vibration signal and drive motor current signal fragment from the online monitoring data of converter transformer OLTC,an adaptive separation method based on endpoint detection is studied in this paper.Firstly,the basic principle of endpoint detection method was introduced,and the separation process of vibration signal and motor current signal was proposed.Then,an OLTC online monitoring system was built and deployed in UHV converter station.Finally,the time-domain and frequency-domain characteristics of online monitoring data was analyzed, and the vibration signal and motor current signal were extracted by differentialendpoint detection technology. Three working stages of the driving motor were divided accordingto the start and end points of the signals. The results show that the endpoint detection method adopted inthis paper can effectively separate OLTC vibration and current motor signals, and the positioning error isless than 5 ms, which can provide a basis for OLTC online monitoring data analysis and fault diagnosis.

关 键 词：换流变压器 有载分接开关 端点检测 振动信号 电机电流信号 自适应分离 

分 类 号：TM315[电气工程—电机]