基于故障传播的模块化BP神经网络电路故障诊断

doi:10.11772/j.issn.1001-9081.2017061516

计算机应用 ›› 2018, Vol. 38 ›› Issue (2): 602-609.DOI: 10.11772/j.issn.1001-9081.2017061516

• 应用前沿、交叉与综合 • 上一篇下一篇

基于故障传播的模块化BP神经网络电路故障诊断

何春, 李琦, 吴让好, 刘邦欣

电子科技大学通信与信息工程学院, 成都 610054

收稿日期:2017-06-20 修回日期:2017-08-14 出版日期:2018-02-10 发布日期:2018-02-10
通讯作者: 李琦
作者简介:何春(1972-),女,四川成都人,副研究员,硕士,主要研究方向:通信信号处理、电子系统的可靠性;李琦(1993-),男,山东滨州人,硕士研究生,主要研究方向:电子系统的可靠性;吴让好(1993-),男,湖北荆州人,硕士,主要研究方向:电子系统的可靠性;刘邦欣(1993-),男,江西赣州人,硕士研究生,主要研究方向:电子系统的可靠性。

Diagnosis of fault circuit by modularized BP neural network based on fault propagation

HE Chun, LI Qi, WU Ranghao, LIU Bangxin

College of Communication and Information Engineering, University of Electronic Science and Technology of China, Chengdu Sichuan 610054, China

Received:2017-06-20 Revised:2017-08-14 Online:2018-02-10 Published:2018-02-10

摘要/Abstract

摘要： 大规模的数模混合电路所含故障模式众多，电路故障状态复杂，且易发生传播，因而电路故障诊断难度较大。针对大规模电路发生故障时存在故障传播的问题，提出一种基于故障传播的模块化BP神经网络（MBPFP）故障诊断方法。首先，在电路模块划分的基础上分析子电路间的故障传播，并将故障源和故障传播源"模块化"；然后，通过子电路的异常检测模型进行一级定位，缩小故障原因集合，确定故障模块；最后，利用目标模块的BP神经网络模型进行二级定位，实现故障诊断并识别故障模式。与传统BP神经网络等方法进行比较的实验结果表明，MBPFP故障诊断方法具有较高的故障覆盖率，在定位准确率方面提高了至少8个百分点，其性能优于传统BP神经网络等方法。

关键词: 大规模数模混合电路, 故障诊断, 故障传播, BP神经网络, 异常检测模型

Abstract: It is difficult to diagnose the faults of large-scale digital-analog hybrid circuit because it has numerous fault modes, the circuit failure status is complex and can be propagated easily. To solve these problems, a new failure diagnosis method, namely Modularized Back Propagation (BP) neural network based on Fault Propagation (MBPFP), was proposed. Firstly, fault propagation between subcircuits was analyzed on the basis of circuit module division, and failure source and transmission source were modularized. Secondly, the set of fault causes was narrowed and the fault module was determined by the anomaly detection model of subcircuit in 1-order positioning. Finally, the fault location was realized and the fault mode was identified by the BP neural network of target module in 2-order positioning. The experimental results show that compared with the traditional BP neural network method, the proposed MBPFP method has a high fault coverage and the accuracy is improved by at least 8 percentage points, which is outperforms the traditional method based on BP neural network.

Key words: large-scale digital-analog hybrid circuit, fault diagnosis, fault propagation, Back Propagation (BP) neural network, anomaly detection model

中图分类号:

TP277.3

何春, 李琦, 吴让好, 刘邦欣. 基于故障传播的模块化BP神经网络电路故障诊断[J]. 计算机应用, 2018, 38(2): 602-609.

HE Chun, LI Qi, WU Ranghao, LIU Bangxin. Diagnosis of fault circuit by modularized BP neural network based on fault propagation[J]. Journal of Computer Applications, 2018, 38(2): 602-609.

参考文献

[1] 谭阳红,何怡刚.模拟电路故障诊断的新故障字典法[J].微电子学,2001,31(4):252-253. (TAN Y H, HE Y G. A new fault dictionary method for fault diagnosis of analog circuits[J]. Microeletronics, 2001, 31(4):252-253.)
[2] 陈文豪,郭子彦,王立,等.复杂机载电子系统故障综合诊断技术研究[J].计算机测量与控制,2016,24(11):1-4. (CHEN W H, GUO Z Y, WANG L, et al. The integrated fault diagnosis technology research based on complex airborne electronic system[J]. Computer Measurement & Control, 2016, 24(11):1-4.)
[3] DUHAMEL P, RAULT J. Automatic test generation techniques for analog circuits and systems:a review[J]. IEEE Transactions on Circuits and Systems, 1979, 26(7):411-440.
[4] 黄亮.模拟电路故障诊断研究[D].北京:北京交通大学,2012:21-24. (HUANG L. Research on fault diagnosis of analog circuits[D]. Beijing:Beijing Jiaotong University, 2012:21-24.)
[5] 何怡刚,罗先觉,邱关源.模拟电路故障诊断L1估计及其神经网络解法[J].电子科学学刊,1997,15(4):365-371. (HE Y G, LUO X J, QIU G Y. A neural-based nonlinear L1 norm optimization algorithm for diagnosis of networks[J]. Journal of Electronics, 1998, 15(4):365-371.)
[6] 袁海英.基于时频分析和神经网络的模拟电路故障诊断及可测性研究[D].电子科技大学,2006:25-49. (YUAN H Y. Study on fault diagnosis & testability in analog circuits based on time-frequency-domain analysis and neural network[D]. Chengdu:University of Electronic and Technology of China, 2006:25-49.)
[7] 王承.基于神经网络的模拟电路故障诊断方法研究[D].成都:电子科技大学,2005. (WANG C. Research on fault diagnosis in analog circuits based on neural network[D]. Chengdu:University of Electronic and Technology of China.)
[8] AMINIAN F, AMINIAN M, COLLINS H W. Analog fault diagnosis of actual circuits using neural networks[J]. IEEE Transactions on Instrument & Meansurement, 2002, 51(3):544-550.
[9] 谭阳红,何怡刚,陈洪云,等.大规模电路故障诊断神经网络方法[J].电路与系统学报,2001,6(4):25-29. (TAN Y H, HE Y G, CHEN H Y, et al. The method of large scale circuit diagnosis based on neutral network[J]. Journal of Circuits and Systems, 2001, 6(4):25-29.)
[10] 张立永.基于数据挖掘的电路故障分析[D].成都:电子科技大学,2015:41-90. (ZHANG L Y. Fault analysis of circuits based on data mining[D]. Chengdu:University of Electronic and Technology of China, 2015:41-90.)
[11] 曾希雯.基于客观化FMEA方法的故障字典研究与实现[D].成都:电子科技大学,2016:28-64. (ZENG X W. Research and implementation of fault dictionary based on objective FMEA[D]. Chengdu:University of Electronic and Technology of China, 2016:28-64.)
[12] FUJIWARA H. FAN:a fanout-oriented test pattern generation algorithm[C]//ISCAS 1985:Proceedings of the 1985 IEEE International Symposium on Circuits and Systems. Piscataway, NJ:IEEE, 1985:671-674.
[13] TAN P N, STEINBACH M, KUMAR V.数据挖掘导论[M].范明,范宏建,译.北京:人民邮电出版社,2014:403-420. (TAN P N, STEINBACH M, KUMAR V. Introduction to Data Mining[M]. FAN M, FAN H J, translated. Beijing:Posts and Telecom Press, 2014:403-420.)
[14] 王健,靳奉祥,史玉峰.模式识别中特征选择与聚类分析[J].测绘工程,2002,11(2):21-24. (WANG J, JIN F X, SHI Y F. Character selection in pattern recognition and cluster analysis[J]. Engineering of Surveying and Mapping, 2002, 11(2):21-24.)
[15] 林超.结合异常检测算法的轴承故障检测研究[D].杭州:浙江大学,2017:59-78. (LIN C. Research on bearing fault detection based on anomaly detection algorithm[D].Hangzhou:Zhejiang University, 2017:59-78.)
[16] 陈侃,李昌禧.故障传播有向图的故障定位研究[J].自动化仪表,2011,32(4):14-17. (CHEN K, LI C X. Research on the fault localization based on directional graphic of fault propagation[J]. Process Automation Instrumentation, 2011, 32(4):14-17.)
[17] 焦李成.神经网络系统理论[M].西安:西安电子科技大学出版社,1996:52-66. (JIAO L C. Neutral Network System Theory[M]. Xi'an:Xidian University Press, 1996:52-66.)
[18] DAVIES D L, BOULDIN D W. A cluster separation measure[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1979, 1(2):224-227.

基于故障传播的模块化BP神经网络电路故障诊断

Diagnosis of fault circuit by modularized BP neural network based on fault propagation

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	孙淑光, 周琪. 基于残差决策的自动飞行控制系统远程实时故障诊断[J]. 计算机应用, 2020, 40(5): 1522-1528.
[2]	王日俊, 白越, 曾志强, 段能全, 党长营, 杜文华, 王俊元. 基于自适应观测器的四旋翼无人飞行器多传感器故障诊断方法[J]. 计算机应用, 2018, 38(9): 2735-2741.
[3]	马梦莹, 曾雅丽, 魏甜甜, 陈志德. 基于邻居节点预状态的无线传感器网络故障诊断算法[J]. 计算机应用, 2018, 38(8): 2348-2352.
[4]	王岩, 罗倩, 邓辉. 基于Gibbs抽样的轴承故障诊断方法[J]. 计算机应用, 2018, 38(7): 2136-2140.
[5]	陈万志, 李东哲. 结合白名单过滤和神经网络的工业控制网络入侵检测方法[J]. 计算机应用, 2018, 38(2): 363-369.
[6]	徐冰珂, 周宇喆, 杨茂林, 谢远航, 李晓瑜, 雷航. 面向电信行业网络告警系统的告警过滤算法[J]. 计算机应用, 2018, 38(10): 2881-2885.
[7]	王军, 费凯, 程勇. 基于改进的Adaboost-BP模型在降水中的预测[J]. 计算机应用, 2017, 37(9): 2689-2693.
[8]	杨洪柏, 张宏利, 刘树林. 基于可视化异类特征优选融合的滚动轴承故障诊断[J]. 计算机应用, 2017, 37(4): 1207-1211.
[9]	石峰, 楼文高, 张博. 基于灰狼群智能最优化的神经网络PM_2.5浓度预测[J]. 计算机应用, 2017, 37(10): 2854-2860.
[10]	张敏龙, 王涛, 王旭平, 常红伟, 王放. 分步动态自回归核主元分析及其在故障诊断中应用[J]. 计算机应用, 2016, 36(5): 1464-1468.
[11]	赵灿明, 李祝红, 陶磊, 张信明. 基于故障传播模型与监督学习的电力通信网络故障定位[J]. 计算机应用, 2016, 36(4): 905-908.
[12]	刘军, 白雪. 基于梯度方向直方图与高斯金字塔的车牌模糊汉字识别方法[J]. 计算机应用, 2016, 36(2): 586-590.
[13]	刘秋玥, 程勇, 王军, 钟水明, 徐利亚. 基于累积和控制图的分布式传感网络故障诊断[J]. 计算机应用, 2016, 36(11): 3016-3020.
[14]	叶卫东, 杨涛. 融合极点对称模态分解与时频分析的单通道振动信号盲分离方法[J]. 计算机应用, 2016, 36(10): 2933-2939.
[15]	许有才, 万舟. 基于条件局部均值分解与变量预测模型的轴承故障诊断方法[J]. 计算机应用, 2015, 35(9): 2606-2610.