基于多尺度熵的滚动轴承Elman神经网络故障诊断方法

张龙; 张磊; 熊国良; 周继惠; 王宁; 王明翔

doi:10.13433/j.cnki.1003-8728.2014.1219

基于多尺度熵的滚动轴承Elman神经网络故障诊断方法

doi: 10.13433/j.cnki.1003-8728.2014.1219

1.
华东交通大学机电工程学院, 南昌 330013

基金项目:

国家自然科学基金项目(51205130,51265010)

江西省教育厅科技项目(GJJ12318)

华东交通大学校立科研基金项目(13JD07)

江西省自然科学基金项目(20132BAB216029)

江西省研究生创新专项基金项目(YC2014-S244,YC2014-S265)资助

详细信息

作者简介:
张龙(1980- ),讲师,博士,研究方向为机械故障诊断,longzh@126.com。

计量
- 文章访问数: 200
- HTML全文浏览量: 46
- PDF下载量: 6
- 被引次数: 0
出版历程
- 收稿日期: 2013-12-02
- 刊出日期: 2014-12-05

Rolling Bearing Fault Diagnosis Based on Multiscale Entropy and Elman Neural Network

1.
School of Mechatronic Engineering, East China Jiaotong University, Nanchang 330013

摘要

摘要: 针对滚动轴承故障振动信号具有跨尺度复杂性的特点,提出了一种新的基于多尺度熵(multiscale entropy,MSE)和反馈式Elman神经网络的滚动轴承故障诊断方法.该方法利用MSE对滚动轴承不同健康状态下的振动信号进行故障特征提取,并将其作为Elman神经网络的输入,利用Elman神经网络自动识别轴承所属的故障类型及故障程度.实验数据包括不同故障类型和不同故障程度样本,结果表明提出的方法能有效地实现滚动轴承故障类型以及程度的智能诊断,效果优于前馈式概率神经网络(Probabilistic neural network,PNN),并具有较低的虚警率和漏警率.
- 多尺度熵 /
- Elman神经网络 /
- 滚动轴承 /
- 故障诊断
Abstract: Considering rolling bearing vibration signals exhibit varied complexity in a multiscale form,a novel rolling element bearing fault diagnosis method was proposed based on multiscale entropy(MSE) and feedback Elman neural network. In this method,the feature extraction from rolling bearing vibrations was performed by utilizing the MSE,and then the MSE was input to the Elman neural network as feature vectors for fault detection.The Elman network was exploited to simultaneously judge bearing fault type and severity degree. Experimental data were collected from an induction motor bearing involving various fault types and severity levels. The results verify the effectiveness of the proposed approach in the sense that the Elman network results in decreased false-alarm and missed-alarm rates in comparison with the feed-forward probabilistic neural network(PNN).
- artificial intelligence /
- backpropagation algorithms /
- efficiency /
- eigenvalues and eigenfunctions

HTML全文

参考文献(14)

[1]	蒋宇,李志雄,唐茗LVQ神经网络在滚动轴承故障诊断中的应用研究[J]机械科学与技术,2011,30(3); 408-411 Jiang Y,Li Z X,Tang M.A Learning vectaor quantization neural network for faults diagnosis of rolling bearing and it's application [J].Mechanical Science and Technology for Aerospace Engineering,2011,30(3): 408-411(in Chinese)
[2]	徐玉秀,原培新,邢钢极大嫡谱法及其在滚动轴承故障诊断中的应用[J]机械科学与技术,2001,20(4); 576-579 Xu Y X,Yuan P X,Xing G.An application of maximum entropy method to fault diagnosis of rolling bearings[J]Mechanical Science and Technology for Aerospace Engineering,2001,20(4):576-579(in Chinese)
[3]	王楠,陈长征,孙长城,等低速滚动轴承故障诊断方法研究[J]机械科学与技术,2007,26[9):llos-llo8 Wang N,Chen C Z,Sun C C,et al.Study of fault diagnosis method for low-speed rolling bearings[J] Mechanical Science and Technology for Aerospace Engineening,2007,26(9):110s-1108(in Chinese)
[4]	张超,陈建军基于LMD近似嫡和支持向量机的轴承故障诊断[J]机械科学与技术,2012,31(9):1539-1542 Zhang C,Chen J J.Diagnosing faults of bearing with LMD approximate entropy and SVM[J]M eehanieal Seience and Technology for Aerospace Engineering, 2012,31(9):1539-1542(in Chinese)
[5]	Pincus S M.Approximate entropy as a measure of system complexity[J]Proe Natl Aead Sei USA,1991,88: 2297-2301
[6]	Richman J S,Moornan J R.Physiolo[cal time series analysis using approximate entropy and sample entropy[J] American Journal of Physiology Heart and Circulatory Physiology,2000,278 (6):H2039-H2094
[7]	Costa M,Goldberger A L,Peng C K.Multiscale entropy analysis of biological signals[J]Physical Review E, 2005,(71):1-18
[8]	Costa M,Goldberger A L,Peng C K.Multiscale entropy analysis of complex physiologics time series[J].Physical Review Letter,2002,89(6):1-18 李冬辉,刘浩基于概率神经网络的故障诊断方法及应用[J]系统工程与电子技术,2004,26(7);998-999 Li D H,Liu H.Method and application of fault diagnosis based on probabilistic: neural network[J] Systems Engineering and Electaronict、,2004,26(7):998-999(in Chinese)
[9]	杨超,王志伟基于Elman神经网络的滚动轴承故障诊断方法[J]轴承,2010,(5);49-52 Yang C,Wang Z W Fault diagnosis method of rolling bearing based on Elman neural network[J]Bearing, 2010,(5);49-52(in Chinese)
[10]	汤宝平,习建民,李锋基于Elman神经网络的旋转机械故障诊断[J]计算机集成制造系统,2010,16 (10);2148-2152 Tang B P,Xi J M,Li F.Fault diagnosis for rotating machinery based on Elman neural network[J] Computer Integrated Manufactauring Systems,2010,16 (10);2148-2152(in Chinese)
[11]	Chen X M,Hong J,Yan H H,et al.Study on error correctaion in dual galvanometer、。arming system based on Elman recurrent neural network [J].Journal of Xi'an Jiaotong University,2006,40(5);587-590
[12]	Case Western Reserve University.Bearing data center [EB/OL][2014-1-6].http;//ctae groups.case.edu/ bearingdata center/pages/wek:ome-case-western-reserve-university-bearing-data-center-website
[13]	夏勇,商斌梁,张振仁,等基于时序分析与神经网络的气阀机构故障诊断[J]机械设计与研究,2001,17 (1);71-72 Xia Y,Shang F L,Zhang Z R,et al.Fault di-agnosis of valve train by using time-series analysis and neural networks[J]Machine Design and Research,2001,17 (1);71-72(in Chinese)
[14]	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[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 inforniation b anulation[J] Mechanical Seience and Technology for Aerospace Engineering,2012,31(1);49-52(in Chinese)