| 摘要 |
PROBLEM TO BE SOLVED: To enable diagnosis even in operation of an apparatus without discontinuing of the operation of an oil-filled electric apparatus and enable highly precise degradation diagnosis even with little learned data. SOLUTION: The method diagnoses the degradation of an oil-filled electric apparatus by estimating the average degree of polymerization of insulating papers in the apparatus from measured values of ingredient amounts of degradation indexes contained in an insulation oil. An input factor group includes each measured value of an furfural amount in the oil, amounts of carbon dioxide and carbon monoxide, an amount of moisture, an amount of oxygen, an amount of hydrogen; the operating history and the maintenance history of the apparatus; all or a part of design specifications of the apparatus. An output factor is the average degree of polymerization of the papers. By identification or learning of models, a plurality of different models for estimating the average degree of polymerization are constructed. The input factor group of the apparatus of an object to be diagnosed is input to the models for estimating and a plurality of estimated values of the average degree of polymerization that are obtained are processed, and consequently the definitive average degree of the papers are estimated. COPYRIGHT: (C)2007,JPO&INPIT
|
