Characteristics of Tin Oxide Chromatographic Detector for Dissolved Gases Analysis of Transformer Oil

Jingmin Fan,Anbo Meng,Feng Bin,Zhe Liu,Qinji Jiang,Hao Yin,Qiuqin Sun

doi:10.1109/access.2019.2928869

Abstract

The analysis of dissolved gases in insulation oil is of great significance to transformer status evaluation. In this paper, a chromatographic detector based on the nano-tin oxide fiber as well as a chromatography system is developed. The mechanism of the sensor for detecting six component feature gases (i.e., H 2 , CO, CH 4 , C 2 H 4 , C 2 H 6 , and C 2 H 2 ) in transformer oil is expounded, on basis of which the exponent-logarithmic model between conductance and gas concentrations is proposed. Then, the repeatability and accuracy of the nano-tin oxide detectors are tested. The experimental results show that the gases mixture can be separated well by the designed gas chromatography system, and six component gases mixture detection can be realized by the developed detector. Meanwhile, by using the proposed model, high precision of dissolved gas measurement can be achieved, thus the validity of the presented model is verified. Moreover, compared with other chromatographic detectors, i.e., flame ionization detector, the only carrier gas needed for the nano-tin oxide chromatographic detector is the synthetic air, the hardware cost and complexity of system are reduced largely, showing promising applicable value in the engineering practice.

Highlights

In the past decades, with the construction of intelligent power grid [1], as well as the digitalization and informatization of substations [2], [3], the periodical maintenance of power transformer will be replaced by the condition based maintenance gradually [4], [5]
Dissolved gas analysis (DGA) of transformer oil is one of the most important methods to detect the latent fault of oil-immersed power transformers [8]
It is worth mentioned that the accuracy and stability of the measuring instrument are affected by gas sensors greatly [9]–[12]

Summary

Introduction

With the construction of intelligent power grid [1], as well as the digitalization and informatization of substations [2], [3], the periodical maintenance of power transformer will be replaced by the condition based maintenance gradually [4], [5]. To promote the performance of SnO2 detector further, in this work, the detection characteristics of nanometer tin oxide sensor are studied; a quantitative mathematical model on basis of exponent-logarithmic model is constructed.

Results

Conclusion