基于遗传BP网络的模拟电路故障诊断

王浩天; 单甘霖; 段修生

doi:10.13433/j.cnki.1003-8728.2014.1226

基于遗传BP网络的模拟电路故障诊断

doi: 10.13433/j.cnki.1003-8728.2014.1226

1.
中国人民解放军军械工程学院, 石家庄 050001

基金项目:

国防科技预研基金重点项目(9140A270202)

自然科学基金项目(51207167)资助

详细信息

作者简介:
王浩天(1989- ),硕士研究生,研究方向为电子系统检测与故障诊断,导航和制导与控制,oec_wht@126.com。

通讯作者:
单甘霖,教授,博士生导师,oec_sg1@126.com。

计量
- 文章访问数: 134
- HTML全文浏览量: 21
- PDF下载量: 4
- 被引次数: 0
出版历程
- 收稿日期: 2013-03-20
- 刊出日期: 2014-12-05

Fault Diagnosis Based on Genetic Algorithm-optimized BP Networks

1.
Mechanic Enhineerinh College, Shijiazhuang 050001

摘要

摘要: 在模拟电路故障诊断中,BP(back propagation)神经网络得到了广泛的应用并取得了不错的效果.但是BP神经网络在训练时仍然存在网络学习收敛速度慢、不易获得全局最优解、网络结构不确定等缺点.采用Levenberg-Marquardt算法进行网络训练,并用遗传算法对BP神经网络结构、初始连接权值和阈值进行全局优选,可以有效克服BP网络存在的缺陷.以Leap Frog Filter滤波器电路的故障诊断为例,仿真实验表明,优化后的BP网络能够快速有效的诊断电路中存在的故障,并且具有更高的诊断精度.
- BP网络 /
- 遗传算法 /
- Levenberg-marquardt算法 /
- 故障诊断
Abstract: Back Propagation(BP) networks have been widely used in analog circuit diagnosis and have achieved some success. However,there are some inherent disadvantages in traditional BP neural network,such as the low speed of error convergency,easily falling into local minimum and the uncertainty structure of networks. Therefore,a new BP network method optimized by genetic algorithms(GA) and Levenberg-Marquardt(LM) algorithm is proposed. In this method,the structure of BP network is optimized by GA. Then LM algorithm is used to train the BP network. The training results can be used to diagnose the faults of analog circuits,which is able to overcome the inherent disadvantages of traditional BP network. The fault diagnosis of the Leap Frog Filter circuit is taken as the example and the simulation results demonstrate the effectiveness and applicability of the proposed method。
- backpropagation algorithms /
- BP networks /
- computer simulation /
- convergence of numerical methods

HTML全文

参考文献(9)

[1]	孙必伟,潘强模拟电路故障诊断的BP神经网络方法研究[J]现代电子技术,2011,34(14);148-153 Sun BW,Pan Q.Research on BP neural network method of fault diagnosis for analog circuits[J]Modern Electaronicta Technique,2011,34(14):148-153(in Chinese)
[2]	吴永新,池阿妮,徐晓辉,等基于BP神经网络的模拟电路故障诊断[J]现代电子技术,2009,32(13);18-20 Wu Y X,Chi A N,Xu X H,et al.Fault diagnosis of analog circuit based on BP neural network[J]Modern Electaronicta Technique,2009,32(13):18-20(in Chinese)
[3]	邓颖,何怡刚,Y.Sun容差模拟电路故障诊断BP神经网络算法[J]湖南大学学报,2000,27(2);56-64 Deng Y,He Y G,Sun Y.Fault diagnosis of analog circuits with tolerances using back-propagation neural networks[J]Journal of Hunan University(Natural Sciences Edition),2000,27(2);56-64(in Chinese)
[4]	Park H I,Park B,Kim Y T,et al.Settlement predictaion in a vertical drainage-installed soft ctaay deposit using GA ack-analysis[J].Marine Georesources&Geotechnology, 2009,27(1):17-33
[5]	李松,刘力军,解永乐遗传算法优化BP神经网络的短时交通流混沌预测[J]控制与决策,2011,26(10); 1581-1585 Li S,Liu L J,Xie Y L.Chaotic: predictaion for short-term traffic: flow of optimized BP neural network based on genetic: algorithm[J]Control and Decision,2011,26 (10):1581-15 85(in Chinese)
[6]	Aminian M,Aminian F.A modular fault-diagnostic system for analog electaronic circuits using neural networks with wavelet transform as a preprocessor[J[.IEEE Trans on Instrumtation Measurement,2007,56(5):1546-1554
[7]	Gu X J,Yao Z T.Fault diagnosis of analog circuits based on improved BP network[J].Electaronic Test,2011,68-11
[8]	Hagan M T,Menbaj M.Training feedforward networks with the levenberg-marquardt algorithm[J]IEEE Transactaions on Neural Networks,1994,5(6);295-301
[9]	张蕾,周洲基于小波和信息粒化的BP神经网络的轴承故障诊断[J]机械科学与技术,2012,31(1);49-52 Zhang L,Zhou Z.The BP neural network fault diagnosis of bearings based on wavelet and information granulation [J Mechanical Seience and Technology for Aerospace Engineering,2012,31(1);49-52(in Chinese)