| Summary: | 碩士 === 國立勤益科技大學 === 電機工程系 === 105 === Large-scale power transformer is one of the most important electrical equipment in the power system. Operating condition affects the power system’s safety and stability directly. Once it is out of function, it will have a big impact and property loss for the whole system and production line. Furthermore, the power system’s safety and stability play significant role via the transformer failure mode research.
To stop the power failure test is no need for dissolved gas analysis in order to facilitate online monitoring. Therefore, it is officially recognized as the oil-filled power transformer that is one of the most effective methods in the early potential failure stage. The study aims to China Steel and Dragon Steel Corporations oil-filled power transformer to evaluate the deterioration performance of the insulating oil.
Finally, by utilizing the neural network and the extension method is to create the diagnosis system. The recognition precision could achieve accurate evaluation result exclude the noise interference according the sample testing result from the neural network diagnosis in comparison to the actual failure type. During the extension diagnosis and go through the extension factor way to figure out the practice of environment, the tolerance of accuracy and temperature variation to find out the evaluation results.
|
|---|
