摘要
由变压器故障诊断中的气体分析法出发，建立了基于人工神经网络中ＢＰ算法的网络模型，并用ＶＢ５．０将其“集成”于一软件包内．测试结果表明，该软件不仅能准确地诊断出５种变压器故障，而且具有非常好的人机交互界面，使对人工神经网络“一无所知”的人员也易于操作，从而为进一步开发变压器在线检测大系统奠定了一定的基础．同时，通过软件的实现也证实可视化语言ＶＢ确实非常适合开发工程类软件．
Transformer fault diagnosis BP artificial neural network is set up based on gas analysis method.It is integrated into a software package using VB50.Five kinds of fault can be diagnosed with the software.The software has a friend interactive interface,and someone not knowing artificial neural network can easily operate it.The developmetn of the software sets up a basis for the further study of large transformer online monitoring systems,and verifies that visual language VB is suitable to developing engineering softwares.
出处
《西安石油学院学报（自然科学版）》
1999年第4期60-63,共4页
Journal of Xi'an Petroleum Institute（Natural Science Edition)
关键词
变压器
神经网络
故障分析
软件
transformer
nerve network
fault analysis
program
automatic system
