BP神经网络在油浸式变压器过热性故障诊断中的应用  

Application of BP Neural Network to Overheating Fault Diagnosis for Oil-immersed Transformer

作　　者：曾成碧[1] 蒲维[2] 曾先锋[3] 

机构地区：[1]四川大学电气信息学院,四川成都610065 [2]四川电力科学研究院,四川成都610072 [3]西昌电业局,四川西昌615000

出　　处：《四川电力技术》2012年第4期60-63,共4页Sichuan Electric Power Technology

摘　　要：针对充油变压器绝缘故障诊断的三比值法的局限性,建立以变压器油中溶解气体含量为样本数据,对不同的隐含层数目进行仿真分析,通过比较确定了适用于变压器绝缘故障诊断的BP神经网络模型。研究表明,这种方法提高了神经网络的收敛速度,符合电力变压器故障诊断系统的实际情况,准确率高。BP neural network model for transformer insulation fault diagnosis is established, according to the limitations of three - ratio method used for the insulation fault diagnosis of oil - immersed transformer. The model is based on the sample data of the content of dissolved gas in transformer oil, and the analysis is based on the number of hidden layers. The results shows that the proposed method can improve the convergence rate of neural network with higher accuracy, which accords with the ac- tual situation of the fault diagnosis system of power transformer.

关 键 词：油浸式变压器 过热性故障 油中溶解气体分析 神经网络 

分 类 号：TM406[电气工程—电器]