New Research Of Transformer Fault Diagnosis Based On Electric Quantity

Power transformer equipment is one of the most critical power systems and is an important guarantee for safe, reliable, high-quality, economic operation to power system. We must maximize to prevent and reduce the occurrence of accidents of transformer. Fault diagnosis and online monitoring techniques to transformer must to be accurate and timely internal transformer fault in order to transformer maintenance timely and reasonable arrangements to ensure reliable power supply.Currently, the diagnostic offline transformer fault mainly based on the original preventive tests and also considering dissolved gas chromatography analysis gradually transit to the transformers online monitoring.。 Windings of the transformer failure are the most parts of transformer failure and those are developed from turn-to-turn short circuit in power transformers.The paper first reviews the status of the transformer fault diagnosis, and then does a detailed analysis to short-circuit transformer turns, finally proposed a new transformer fault diagnosis theory and test the theory with the new dynamic simulation test. The method is the use of transformer winding turn fault of the changes and abrupt loss for fault diagnosis, According to the new electrical features: loss ratio of negative-sequence combined the loss and negative-sequence loss transformers, it can determine whether there is turn transformer short-circuit and fault phase. Through theoretical analysis and proved that the project can be applied to the actual transformer fault diagnosis and the method of calculating the threshold can be applied to the fault diagnosis. In Dynamic Simulation Lab with its associated electrical dynamic simulation test the experimental results indicate that this method is more progress than transformer loss method of fault diagnosis transformer. Integration of the transformer historical data and the transformer oil Chromatographic analysis of the transformers, the method will be available online to raise the level of detection.