基于流形子带特征映射的转子复合故障特征提取方法

王广斌1 李龙1 罗军 2 杜晓阳3 李学军1

振动与冲击 ›› 2017, Vol. 36 ›› Issue (16) : 56-62.

PDF(858 KB)
PDF(858 KB)
振动与冲击 ›› 2017, Vol. 36 ›› Issue (16) : 56-62.
论文

基于流形子带特征映射的转子复合故障特征提取方法

  • 王广斌1  李龙1 罗军 2 杜晓阳3 李学军1
作者信息 +

Rotor Compound Fault Feature extraction Based on Manifold subband feature mapping method

  • Wang Guangbin1 Li Long1 Luo Jun 2 Du xiaoyang3 Li xuejun1
Author information +
文章历史 +

摘要

针对复合故障特征易被噪声信号淹没,传统时频分析和流形学习方法不能完整有效的挖掘故障潜在信息和进一步实现故障特征提取。本文在流形学习的基础上提出了一种流形子带思想并将其应用到转子复合故障特征提取研究中,进而得出了一种基于流形子带特征映射的转子复合故障特征提取方法。首先对故障原始信号序列进行相空间重构,结合小波包对噪声的强烈抑制性和对信号分辨率高的特点,将重构信号分解成不同频带即子带。然后将同故障多种工况下的同一频带融合成频带矩阵并估计其本征维数,最后通过拉普拉斯特征映射算法以本征维数为依据将子带降维获取低维特征向量并提取信息熵,进一步实现故障特征提取。实验表明,相对于经典的局部线性嵌入和拉普拉斯特征映射等算法,流形子带特征映射算法不仅对单故障而且对复合故障特征进行了更完整有效的挖掘和提取。

Abstract

For compound fault features easy submerged by noise signal. Traditional time-frequency analysis and manifold learning method can't complete failure for efficient mining of potential information and further realize the fault feature extraction. Based on manifold learning is proposed on the basis of a manifold subband thought and its application to the study of compound rotor faults. Then draw a manifold subband feature mapping method based on compound rotor fault. First of all,to the phase space reconstruction fault original signal sequence. combination with wavelet packet strong inhibitory to noise and the signal the characteristics of high resolution. Reconstructing signal is decomposed into different frequency bands. The same frequency band of the same fault and many conditions are intergrated into the band matrix and estimate the intrinsic dimension. At last, by Laplace feature mapping algorithm based on the intrinsic dimension Subband dimension reduction obtained low dimensional feature vector and extract the information entropy, then further realized the fault feature extraction. Experiments show that: Compared with the classical local linear embedding and Laplace feature map algorithm, etc. Not only for single fault but also for compound fault manifold subband feature mapping algorithm more completely and effectively digged and extracted the characteristics.
 

关键词

转子系统 / 流形子带 / 拉普拉斯特征映射算法 / 特征提取

Key words

Rotor system / Manifold sub-band / Laplacian Eigenmaps / Feature extraction

引用本文

导出引用
王广斌1 李龙1 罗军 2 杜晓阳3 李学军1 . 基于流形子带特征映射的转子复合故障特征提取方法[J]. 振动与冲击, 2017, 36(16): 56-62
Wang Guangbin1 Li Long1 Luo Jun 2 Du xiaoyang3 Li xuejun1 . Rotor Compound Fault Feature extraction Based on Manifold subband feature mapping method[J]. Journal of Vibration and Shock, 2017, 36(16): 56-62

参考文献

[1]. Seung H S, Daniel D L.The manifold ways of perception[J].Science(S0036-8075),2000,290(5500):2268-2269.
[2]. Roweis S, SaulL. Nonlinear dimensionality reduc-tion by locally linear embedding[J]. Science(S0036-8075),2000,290(5500):2323-2326.
[3]. Belkin M, Niyogi P. Laplacian eigenmaps for dimensionality reduction and data representation[J].NeuralComputation. 2003,15(6):1373-1396.
[4]. Zhang Zhenyue, Zha Hongyuan. Principal Mani-folds and nonlinear dimension reduction viatang-ent Space Alignmnet[J]. SIAM Journal on Scien-tific Computing,2005,26(1):313-338.
[5]. He QB, Liu YB, Long Q.Time-frequency manifo-ld as a signature for machine health diagnosis[J]. IEEE Transactions on Instrumentation and Meas-urement, 2012, 61(5): 1218-1230.
[6]. QingboHe, XiangxiangWang.Time-frequency ma-nifold correlation matching for periodic fault identification in rotating machines[J]. Journal of Sound and Vibration, 2013, 332(10): 2611-2626.
[7]. Qingbo He. Time-frequency manifold for nonlin-ear feature extraction in machinery fault diagnos-is[J]. Mechanical Systems and Signal Processing, 2013, 35(12): 200-218
[8]. Yi Wang, Guanghua Xu, Lin Liang, Kuosheng Jiang. Detection of weak transient signals based on wavelet packet transform and manifold learnin-g for rolling element bearing fault diagnosis[J]. Mechanical Systems and Signal Processing, 2015,54-55(3): 259-276
[9]. Baoping Tang, Tao Song, Feng Li. Fault diagnos-is for a wind turbine transmission system based on manifold learning and Shannon wavelet suportvector machine[J]. Renewable Energy, 2014, 6(2): 1-9.
[10]. 栗茂林, 梁霖, 王孙安, 庄健. 基于连续小波系数非线性流形学习的冲击特征提取方法[J]. 振动与冲击, 2012, 31(01): 106-112.
[11]. 李锋, 汤宝平. 基于线性局部切空间排列维数化简的故障诊断[J]. 振动与冲击, 2012, 31(13): 36-41.
Li Feng,Tang Baoping.Fault diagnosis model bas-ed on dimension reduction using linear local tag-ent space alignment[J].Journal of Vibration and Shock, 2012, 31(13): 36-41.
[12]. 向丹, 葛爽. 基于EMD 样本熵-LLTSA 的故障特征提取方法[J]. 航空动力学报,2014, 29(7): 1535-1542.
XIANG Dan,GE Shuang.Method of fault feature extraction based on EMD sample entropy and LLTSA[J].Journal of Aerospace Power, 2012, 31(13): 36-41.
[13]. 刘韬, 陈进, 董广明. 基于频带熵的滚动轴承故障诊断研究[J]. 振动与冲击. 2014, 1(33): 77-80.
LIU Tao,CHEN Jin,DONG Guang-ming. Rolling element bearing fault diagnosis based on frequen-cy band entropy[J].Journal of Vibration and Sho-ck, 2014, 1(33): 77-80.
[14]. 张靖, 闻邦椿. 两端支座松动转子系统的频率特性分析[J].中国机械工程,2008, 19(1): 68-71.
Zhang Jing, Wen Bangchun.A Study of Frequen-cy Characteristics of Rotor Sysrem with Pedestal Looseness at Two Supports[J].Chaina Mechanical Eegineering,2008, 19(1): 68-71.
[15]. N.H.Paekard,J.PCrutehfietd,J.D.Farmer,etal.Geometry from a time series.Phys.Rev.Lett, 1980, 45(9): 712-716.
[16]. F.Takens.Determing strange attrae torsin tuthulen -ee. Leeture notes in Math, 1981, 898: 361-381.
[17]. 周云龙, 张学清, 高云鹏. 基于小波包多尺度信息熵和HMM的气液两相流流型识别方法[J]. 核科学与工程,2009, 29(4): 333-339.
ZHOU Yunlong,ZHANG Xueqing,GAO Yunpen-g.A method for identifying gas-liquid two-phase flow patterns on the basis of wavelet packet multi-scale information entropy and HMM[J].Chinese Journal of Nuclear Science and Engineering, 2009, 29(4): 333-339.
[18]. 陈维省.微分流形初步(第二版)[M].北京:高等教育出版社,2001:2-5.
Chen Weixing. Differential manifold preliminar-y(second edition)[M].Beijing:Higher education press,2001:2-5.
[19]. 孙明明.流形学习理论与算法研究:[D].南京:南京理工大学, 2007:5-12.
Sun Mingming.The theory and algorithm of man-ifold learning:[D].Nanjing:Nanjing University o-f Science and Technology, 2007:5-12.

PDF(858 KB)

Accesses

Citation

Detail

段落导航
相关文章

/