Odor recognition of deteriorated mineral oils using an odor-sensing array

The deterioration or oxidation of the mineral oil in transformers poses risk short circuits. Convenient and effective methods are expected to be developed. Carbon-based sensor arrays were used this study identify quality variations for oil-filled by odors. sensitive layers odor-sensing system consisted different types GC stationary phase materials carbon black (CB) mixtures. We made a targeted selection utilizing polarities make array based on distinct components such as alkanes xylenes gas chromatography mass spectrometry (GC/MS) analysis. response characteristics recognize odors machine learning. With laboratory air carrier gas, could distinguish that has been use over 20 years from new with an accuracy about 93.8%. identification achieved was 60% three concentrations unused oxidized created transformer’s leakage. When detecting concentration more than 50%, rate reached 80%. will help inspect oils transformer leakage judgments time.