Safe Operation Of Power Transformer Line Monitoring And Fault Diagnosis Technology Research

Power transformer is one of the most important equipments in substation, its running condition relates to all the line. For a long time, we have adopted time-based maintenance to transformer, this way of management seriously interferences the line. For recent years, with quickly improving of the power capacity and its automation, the way of time-based maintenance is quite not adapt to the security of the line and the development of the economics.On-line monitoring of transformers with different built-in sensors directly in the transformer, where possible, this should carry as much information as possible on the state of the transformer at a particular moment. Then we can make a proper diagnosis with this information, and take measures to prevent serious faults happening.The insulation of transformer contains liquid insulation (mineral oil) and solid insulation (insulation paper). Mineral insulating oil is made of a blend of different hydrocarbon molecules containing CH₃, CH₂ and CH chemical groups linked together. It may be broken as a result of electrical and thermal faults, in radical or ionic form, into gas molecules such as hydrogen (H₂), methane (CH₄), ethane (C₂H₆), ethylene (C₂H₄), acetylene (C₂H₂). Solid insulation may produce CO, CO₂ with electrical and thermal faults. These gases most dissolve in the oil. If we separate these gases from the oil and analyze them, we can estimate the possibility of fault and the type of fault.With the development of gas-chromatographic diagnostic techniques, the sensor techniques and fault diagnostic techniques, on-line monitoring of transformer oil dissolved gases has developed rapidly. This paper discusses gases dissolved in transformer oil on-line monitoring. Through study of macromolecule membrane and sensor techniques, this paper proposes using multi-sensors to detect six characteristic gases (H₂ CH₄ C₂H₆ C₂H₄ C₂H₂ CO) and combining characteristic gas-based neural network and proved three gas ratios-based fuzzy neural network to analyze the oil dissolved gases. Through examples we can find the method valid.