摘要
针对电子设备故障诊断中数据处理量大,容错能力差,匹配冲突等问题,提出了基于神经网络、证据理论信息融合进行故障诊断的方法;首先,应用BP神经网络将特征提取后的信息进行特征级融合,实现了一定的数据压缩,然后,采用证据理论对不同的诊断结果组成的证据体进行决策级融合,减小故障诊断的不确定性,并解决各故障之间匹配冲突的问题;最后,以某型雷达I/O接口板为例说明了本文方法的有效性和实用性。
In view of the electronic equipment of date processing capacity, poor fault tolerance and the match conflict, an evaluation method of fault diagnosis based on information fusion is proposed. First, BP neural network is used to feature--level fusion, so as to achieve a certain degree of date compression. Then using the evidence theory for decision--level fusion of different evidence body in order to reduce the uncertainty of fault diagnosis and solve the problem of matching conflict between the various faults. Finally, an example of an I/O inter- face board of a radar equipment is provided and the validity and practicability of the method proposed is confirmed.
出处
《计算机测量与控制》
CSCD
北大核心
2012年第11期2888-2890,2893,共4页
Computer Measurement &Control
基金
河北省重点基础研究项目(10963529D)
关键词
信息融合
人工神经网络
证据理论
故障诊断
information fusion
artificial neural network
evidence theory
fault diagnosis
