Study on the development and test method of SnO2-based gas sensor array for dissolved gas analysis

Dissolved Gas Analysis (DGA) is one of the most effective method to diagnose the early latent fault in oil-immersed power transformer. Gas sensor detection technology is the core of DGA, which will directly affect the reliability and accuracy of the transformer condition monitoring and fault diagnosis. At present, since the development of gas sensor restricted by the characteristics of dispersion, low sensitivity, ageing or poisoning, therefore, it is actively significance to continuously study the gas sensing technology and develop new type gas detection sensor to improve the on-line monitoring level of gas dissolved in transformer oil. In this work, gas sensor array was developed by synthesizing the Ag/Zn/Cu/Pt-loaded SnO
₂
hybrid nanocomposite via hydrothermal method. The study of the relationship between input voltage and gas sensitivity, gas selection and the drift characteristics between temperature and humidity to two typical characteristic fault gases hydrogen (H
₂
) and acetylene (C
₂
H
₂
) for as-prepared gas sensor array were carried out. In addition, based on combined intelligent algorithm (PCA-BPNN), the classification and quantitative research of mixed gases were analyzed and discussed. The results laying the foundation for the preparation of gas sensor which serve for the transformer condition monitoring and fault diagnosis.