BP神经网络在舰艇电路系统故障诊断仪中的应用  

Application of BP Neural Network in Vessel Circuits Fault Diagnosis Tester

作　　者：游望星[1] 何怡刚[1] 祝文姬[1] 胡湘娟[1] 

机构地区：[1]湖南大学电气与信息工程学院,湖南长沙410082

出　　处：《计算机测量与控制》2009年第8期1565-1567,1585,共4页Computer Measurement &Control

基　　金：国家863计划项目(2006AA04A104);国家自然科学基金项目(50677014);高校博士点基金(20060532002);湖南省科技计划项目(06JJ2024);教育部新世纪优秀人才支持计划项目(NCET-04-0767)

摘　　要：详细介绍了BP神经网络的原理,并利用训练样本对网络进行了训练,重点针对BP神经网络在舰艇电路系统故障诊断仪中的软硬件实现进行了设计;软件部分采用VB面向对象编程技术和MATLAB快速矩阵运算能力相结合实现BP神经网络的诊断;硬件部分设计以数字信号处理器(DSP)为核心;根据文中的设计方法制作的测试样机通过了技术鉴定,鉴定结果表明该方法具有很好的快速性和精确性,对舰艇电路系统的智能故障诊断技术具有很好的借鉴意义。The theory of BP neural network was detailedly introduced in this paper and the BP neural network is trained through some samples. The focus in this paper are hardware and software designs of BP neural network for vessel circuits fault diagnosis tester. The combination of VB object--oriented programming technology and the rapid MATLAB matrix computing ability was used to achieve BP neural net-work diagnosis in vessel circuits fault diagnosis tester＇ s software design; in hardware design digital signal processor （DSP） was the core. A prototype tester was designed according to the method in this paper and the prototype tester passed technical appraisal. The results have showed the good speed and accuracy of the method, as well as its reference to the intelligent fault diagnosis technology of vessel＇ s circuit system.

关 键 词：舰艇电路系统故障诊断仪 BP神经网络 模拟电路 设计 

分 类 号：TP277[自动化与计算机技术—检测技术与自动化装置]