变压器色谱监测中的 BPNN 故障诊断法  

Monitoring Transformer Chromatography with BPNN Faults Diagnosis Method

作　　者：王财胜[1] 孙才新[1] 廖瑞金[1] 

机构地区：[1]重庆大学电气工程系

出　　处：《中国电机工程学报》1997年第5期322-325,共4页Proceedings of the CSEE

摘　　要：本文将ＢＰ神经网络应用于变压器故障诊断。建立起学习样本集，提出了两种输入方式，并用它对神经网络进行训练。通过验证，结果显示该ＢＰＮＮ诊断法有良好的应用前景。In this paper,that back propagation neural network (BPNN) is applied in transformer fault diagnosis is discussed.The teaching set of the BPNN has been set up with using chromatography records from many substations.Two types of inputting vector for this BPNN are presented and the training bas been made respectively.The results of verification show that the BPNN diagnosis method has a bright application prospect.

关 键 词：变压器 BP神经网络 色谱监测 故障诊断 

分 类 号：TM410.6[电气工程—电器]