首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于紫外光谱检测的GIS内多类故障早期预警
引用本文:赵宇,王先培,胡红红,代荡荡,龙嘉川,田猛,朱国威,黄云光.基于紫外光谱检测的GIS内多类故障早期预警[J].光谱学与光谱分析,2015,35(2):438-442.
作者姓名:赵宇  王先培  胡红红  代荡荡  龙嘉川  田猛  朱国威  黄云光
作者单位:1. 武汉大学电子信息学院,湖北 武汉 430072
2. 深圳供电局有限公司,广东 深圳 518001
3. 广西电力科学研究院,广西 南宁 530015
基金项目:国家自然科学基金项目,中国南方电网科技项目,湖北省科学条件专项基金
摘    要:气体绝缘开关(GIS)故障预警强调异常识别的实效性和现场适用性,是精确诊断的基础。研究GIS内局部放电(PD)及过热故障的统一预警方法具有重要意义。提出将SO2作为GIS内部两类故障公共的特征分解物,并应用紫外光谱(UV)测量SO2实现多故障早期预警。一阶导数法和Savitzky-Golay滤波被用于快速基线校正和光谱平滑,290~310 nm被选为特征区进行SO2检测及定量。UV检测可避免诸如SF6及SOF2、SO2F2等分解物的干扰,便于从复杂的SF6分解物组分中识别微量SO2特征,测量系统结构简单、维护成本低、适于现场检测。通过模拟SF6气体在两类PD缺陷及200~400 ℃过热条件下的分解实验,证实所提UV预警方法的有效性:两类GIS故障条件下都会稳定的生成SO2。对UV预警后的故障诊断,借助傅里叶变换红外光谱仪及气相色谱仪,确认PD及过热故障下的分解物组分存在差异,并据此简要分析了故障类型辨识方法。PD分解物以SOF2和SO2F2为主,SO2含量远小于SOF2,放电涉及环氧树脂时含碳分解物含量上升。过热材质为不锈钢时,SF6约在300 ℃发生分解,300~400 ℃内的过热分解物主要为SO2和SOF2,350 ℃以上SO2生成速率明显快于PD时。随着GIS内故障持续恶化,SO2含量均呈现稳定上升趋势。在UV预警后,可利用获得的SO2生成特性初步辨识故障类型。

关 键 词:气体绝缘开关  局部放电  过热故障  故障统一预警  紫外光谱  
收稿时间:2014-03-28

Early Warning for Various Internal Faults of GIS Based on Ultraviolet Spectroscopy
ZHAO Yu,WANG Xian-pei,HU Hong-hong,DAI Dang-dang,LONG Jia-chuan,TIAN Meng,ZHU Guo-wei,HUANG Yun-guang.Early Warning for Various Internal Faults of GIS Based on Ultraviolet Spectroscopy[J].Spectroscopy and Spectral Analysis,2015,35(2):438-442.
Authors:ZHAO Yu  WANG Xian-pei  HU Hong-hong  DAI Dang-dang  LONG Jia-chuan  TIAN Meng  ZHU Guo-wei  HUANG Yun-guang
Institution:1. School of Electronic Information, Wuhan University, Wuhan 430072, China2. Shenzhen Power Supply Bureau Company, Limited, Shenzhen 518001, China3. Guangxi Grid Electric Power Research Institute, Nanning 530015, China
Abstract:As the basis of accurate diagnosis, fault early-warning of gas insulation switchgear (GIS) focuses on the time-effectiveness and the applicability. It would be significant to research the method of unified early-warning for partial discharge (PD) and overheated faults in GIS. In the present paper, SO2 is proposed as the common and typical by-product. The unified monitoring could be achieved through ultraviolet spectroscopy (UV) detection of SO2. The derivative method and Savitzky-Golay filtering are employed for baseline correction and smoothing. The wavelength range of 290~310 nm is selected for quantitative detection of SO2. Through UV method, the spectral interference of SF6 and other complex by-products, e.g. SOF2 and SOF2, can be avoided and the features of trace SO2 in GIS can be extracted. The detection system is featured by compacted structure, low maintenance and satisfactory suitability in filed surveillance. By conducting SF6 decomposition experiments, including two types of PD faults and the overheated faults between 200~400 ℃, the feasibility of proposed UV method has been verified. Fourier transform infrared spectroscopy and gas chromatography methods can be used for subsequent fault diagnosis. The different decomposition features in two kinds of faults are confirmed and the diagnosis strategy has been briefly analyzed. The main by-products under PD are SOF2 and SO2F2. The generated SO2 is significantly less than SOF2. More carbonous by-products will be generated when PD involves epoxy. By contrast, when the material of heater is stainless steel, SF6 decomposes at about 300 ℃ and the main by-products in overheated faults are SO2 and SO2F2. When heated over 350 ℃, SO2 is generated much faster. SO2 content stably increases when the GIS fault lasts. The faults types could be preliminarily identified based on the generation features of SO2.
Keywords:Gas insulation switchgear  Partial discharge  Overheated fault  Fault unified early-warning  Ultraviolet spectroscopy
本文献已被 CNKI 万方数据 等数据库收录!
点击此处可从《光谱学与光谱分析》浏览原始摘要信息
点击此处可从《光谱学与光谱分析》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号