365 / 2022-03-15 17:26:15

Study on the Detection of Multi-components Gas Based on Metal Doped SnO2 Sensor Array

sensor array,transformer fault,neural network algorithm,nanomaterials

Purpose/Aim

Methane (CH₄) and acetylene (C₂H₂) are two kinds of characteristic gases which can reflect critical overheating and discharge fault in power transformer. Gas sensor array coated with gas sensitive metal doped SnO₂ can separate concentration of CH₄ and C₂H₂ efficiently combining Principal Component Analysis (PCA) and Back Propagation (BP) algorithm.

Experimental/Modeling methods

The structure of sensor array is designed and manufactured by Micro Electro Mechanical System (MEMS). Nano-materials SnO₂ doped with Cu, Zn, Pt and Ni are composited by hydrothermal method and characterized by X-ray diffraction (XRD), transmission electron microscopy (TEM) method. At last, the sensitivity and selectivity are verified and PCA+BP algorithm are utilized to reduce the cross sensitivity.

Results/discussion

Zn-doped SnO_{2 ­}nanomaterial is the most sensitive one to CH₄ and Pt-doped SnO_{2 ­}nanomaterial is the most sensitive one to C₂H₂.  PCA+BP algorithm can separate the concentration of CH₄ and C₂H₂ mixed efficiently compared with outcomes which only utilizes BP algorithm.

Conclusions

The structure of sensor array is manufactured successfully which can have mass production by assembly line. Besides, the sensor array can be put into use in power system in a large scale for reflecting the potential fault in power transformer.