摘要
针对目前指控装备通信设备故障诊断准确率低,诊断方法通用性差的现状,提出一种基于小波神经网络的故障诊断方法。该方法利用了小波分析良好的时频特性和局部变焦能力以及神经网络自学习能力和处理大量数据的能力,同时两者的结合进一步优化了神经网络结构,提高了诊断效率。仿真实验表明,基于小波神经网络的故障诊断方法大幅提高了故障诊断的效率和准确率。
Aiming at the current situation of low fault diagnosis accuracy of command equipment and communication equipment and poor generality of diagnosis methods,a fault diagnosis method based on wavelet neural network was proposed.This method took advantage of the good time-frequency characteristics of wavelet analysis and local zooming ability,as well as the self-learning ability of neural network and the ability to process large amounts of data.At the same time,the combination of the two further optimized the neural network structure and improved diagnosis efficiency.Experiments showed that the fault diagnosis method based on wavelet neural network greatly improved the efficiency and accuracy of fault diagnosis.
作者
陈旭
胡建旺
孙慧贤
单成进
CHEN Xu;HU Jianwang;SUN Huixian;SHAN Chengjin(Shijiazhuang Campus of Army University of Engineering,Shijiazhuang 050003,China;Unit 32228 of PLA,Xiamen 361100,China)
出处
《探测与控制学报》
CSCD
北大核心
2020年第6期55-60,共6页
Journal of Detection & Control
基金
国防预研基金项目资助(41404030101)。
关键词
指控装备
故障检测
小波神经网络
通信控制机测试
accused equipment
communication controller test
wavelet neural network
fault detection