Multi-Gas Detection in Power Transformer Oil Based on Tunable Diode Laser Absorption Spectrum
Multi-Gas Detection in Power Transformer Oil Based on Tunable Diode Laser Absorption Spectrum
复制标题
基于可调谐二极管激光吸收光谱的电力变压器油中多种气体检测
DOI：
10.1109/tdei.2018.007535
复制
发表时间：
2019-02-01
影响因子：
3.1
通讯作者：
Luo, Yingting
中科院分区：
文献类型：
--
作者：
Jiang, Jun;Wang, Zhuowei;Luo, Yingting
在线溶解气体分析(DGA)是获取电力变压器绝缘油健康状态信息的有效技术。绝缘油分解出的典型烃类气体(甲烷、乙炔、乙烯、乙烷)是诊断故障类型和严重程度的关键指标。为了实现非接触式现场测量，提出了一种基于可调谐半导体激光吸收光谱(TDLAS)的多气体检测系统。研究了光源、长光程气室和多气体检测拓扑结构等关键技术问题。在近红外波段，烃类气体的各个中心波长被设定为甲烷为1653.72 nm，乙炔为1530.37 nm，乙烯为1620.04 nm，乙烷为1679.06 nm。研制了一种长10.13m的多通道气室，具有抗振设计的优点。为了共享光吸收路径，采用光开关对多路气体检测系统进行合并。实验室校准证明了TDLAS多气体检测系统的高灵敏度，乙炔的检测达到百万分之几，其他烃类气体的检测达到ppm级。通过在实际220kV电力变压器上的对比检测，证明了DGA监测的有效性，为电力变压器油中多种气体的在线检测提供了一种新的途径。
Online dissolved gas analysis (DGA) is an effective technique to obtain the health status information of insulation oil in power transformers. The typical hydrocarbon gases (methane, ethyne, ethene, ethane) decomposed from insulation oil, are the key indicators to diagnose the fault type and severity. To realize contactless and field measurement, a multi-gas detection system based on tunable diode laser absorption spectrum (TDLAS) is proposed in this paper. Critical technique problems on the light source, long path gas cell, and the topology for multi-gas detection are investigated. Individual central wavelengths of hydrocarbon gases are set at as 1653.72 nm for methane, 1530.37 nm for ethyne, 1620.04 nm for ethene and 1679.06 nm for ethane in the near infrared band. A 10.13 m long multi-pass gas cell is developed with the advantage of anti-vibration design. To share the optical absorption path, an optical switch is adopted to merge the multi-gas detection system. The high sensitivity of the TDLAS multi-gas detection system was demonstrated in the laboratory calibrations, with ethyne detection reaching sub-parts per million level, and the other hydrocarbon gases achieving ppm level. Furthermore, the comparison detection in the real 220 kV power transformer gave evidence to the effective monitoring of DGA, proving it an alternative approach to online detection of multiple gases in power transformer oil.