BP神经网络改进IEC-60599在电力变压器故障诊断中的应用

Application of BP Neutral Net Modified IEC-60599 in Power Transformer′s Fault Diagnosis

由于电力变压器是电力系统中重要的故障原因、故障现象和故障机理间存在着随机性和模糊性等不确定因素,故障特征量和故障征兆间没有明确的一一对应关系,这就给故障诊断增加了困难。BP神经网络诊断变压器故障已经越来越受到重视,可常规的BP算法对大样本数据进行训练时,大多存在收敛速度慢、甚至有时不收敛的缺点,因此提出了改进BP算法从而加快了训练速度和增加了诊断的可靠性。对文献[2]中提出的改进的新导则IEC-60599三比值诊断方法在进行中、低温过热故障时进行修正。

A power transformer is an important fault reason for a power system, some uncertain factors such as randomness and fuzniness exist between fault phenomenon and fanlt mechanism. There are no definite corresponding relations between fault characteristic quantity and fault omens, which make fault diagnosis more difficult. The transformer fault of BP neutral net diagnosis is more and more paid attention to ,but when the normal BP claculations to the data are trained, most of them have slow convergent speed, even no convergent. The paper proposeed to improve BP calculation, thus increase training speed and diagnostic reliability. Three ratio diagnosis method in medium overhot fault and low temperature to improve new guidance IEC -60599 in Doucment 2 has been proposed are revised.