用户名: 密码: 验证码:
基于神经网络的故障诊断及应用
详细信息    查看官网全文
摘要
故障诊断是故障预测和健康管理的主要功能之一,是结构复杂、子系统繁多的原因,一般难以准确、高效地诊断出故障。利用神经网络的非线性处理、分布式并行处理等特性,研究了基于BP神经网络的故障诊断算法,选用某种典型故障征兆数据进行故障诊断仿真,并对相关故障诊断神经网络进行了噪声测式,仿真结果表明,所建立神经网络都能很好地完成故障诊断任务。
As one of the main features of PHM and the most primitive feature,fault diagnosis identifies the type of fault and specifies the location of fault.Due to its complex structures,the fault of avionics systems is generally difficult to diagnose accurately and efficiendy.Using non-linear processing and distributed parallel processing of the neural networks,the fault diagnosis algorithm based on back propagation(BP) neural network is studied.Selected a typical airborne fire control systems' fault symptoms data to do simulation,the noise measurement for these neural network is done.The experimental results show that the fault diagnosis neural network can complete the fault diagnosis task very well.
引文
[1]张凤鸣,惠晓滨.航空装备故障诊断学[M].北京:国防工业出版社,2010.
    [2]杨保安,张科静.多目标决策分析:理论、方法与应用研究[M].上海:东华大学出版社,2008.
    [3]冯琦,肖桥,周德云.基于BP神经网络的机载数字高程模型压缩[J].航空工程进展,2011,2(3):339-343.
    [4]张智星,孙春在,[日]水谷英二.神经-模糊和软计算[M].西安:西安交通大学出版社,2000.

© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号

地址:北京市海淀区学院路29号 邮编:100083

电话:办公室:(+86 10)66554848;文献借阅、咨询服务、科技查新:66554700