基于噪声辅助多元经验模态分解和多尺度形态学的滚动轴承故障诊断方法

武哲1,杨绍普2,任彬2,马新娜2,张建超1,2

振动与冲击 ›› 2016, Vol. 35 ›› Issue (4) : 127-133.

PDF(2627 KB)
PDF(2627 KB)
振动与冲击 ›› 2016, Vol. 35 ›› Issue (4) : 127-133.
论文

基于噪声辅助多元经验模态分解和多尺度形态学的滚动轴承故障诊断方法

  • 武哲1,杨绍普2,任彬2,马新娜2,张建超1,2
作者信息 +

Rolling Element Bearings Fault Diagnosis Method Based on NAMEMD and multi-scale morphology

  • Wu Zhe1,Yang Shao-pu2,Ren Bin2,MA Xinna2,Zhang Jian-chao1,2
Author information +
文章历史 +

摘要

为了从强噪背景中提取滚动轴承微弱故障特征,提出一种基于噪声辅助多元经验模态分解 (Noise Assisted Multivariate Empirical Mode Decomposition,NAMEMD)和数学形态学的滚动轴承故障诊断方法。NAMEMD是新提出的一种基于噪声辅助数据分析方法,其克服了集成经验模态分解的模态混淆和运算量大等问题。本文将NAMEMD与多尺度形态学相结合应用于滚动轴承故障诊断。该方法首先利用NAMEMD将多分量调频调幅故障信号自适应分解为一系列IMF分量;其次,选取能量高的IMF分量求和重构;最后利用多尺度形态学差值滤波器提取信号的故障特征频率。为了验证理论的正确性,进行了仿真试验和轴承故障试验,并与EEMD和包络解调进行了比较,结果表明本文方法在进一步降低模态混叠效应的同时,明显提高了运算速度,对滚动轴承外圈、内圈和滚子故障的检测精度更高,能够清晰地提取出故障信号的故障特征频率。

Abstract

This paper proposes a rolling bearing fault diagnosis method based on multivariate empirical mode decomposition-based noise assisted multivariate empirical mode decomposition (NAMEMD) and mathematical morphology. NAMEMD, as a noise assisted data analysis-based method, could effectively avoid such shortcomings of ensemble empirical mode decomposition as mode mixing and heavy computation, thus being superior to traditional noise assisted data analysis-based method to a certain extent. This paper uses NAMEMD with multiscale morphology for rolling bearing fault diagnosis. As for the method proposed, NAMEMD is used to adaptively decompose multi-component FM and AM fault signal into a series of IMF components, of which the high-energy ones are subjected to summation & reconstruction; then, a multiscale morphological difference filter is employed to extract the fault characteristic frequency of signal. In order to verify the correctness of theory, simulation experiment and bearing fault experiment were performed to compare with EEMD and envelope demodulation-based methods; according to the result, the present technique could further alleviate mode mixing effect, significantly improve the computation speed, bring about higher detection accuracy for the faults in outer race, inner race and roller in rolling bearing, and clearly extract the frequency characteristics of fault signal.
 

关键词

NAMEMD / 模态混叠 / 多尺度形态学 / 滚动轴承 / 故障诊断

Key words

NAMEMD;Mode Mixing;multiscale morphology / Rolling Bearing / Fault diagnosis

引用本文

导出引用
武哲1,杨绍普2,任彬2,马新娜2,张建超1,2. 基于噪声辅助多元经验模态分解和多尺度形态学的滚动轴承故障诊断方法[J]. 振动与冲击, 2016, 35(4): 127-133
Wu Zhe1,Yang Shao-pu2,Ren Bin2,MA Xinna2,Zhang Jian-chao1,2. Rolling Element Bearings Fault Diagnosis Method Based on NAMEMD and multi-scale morphology[J]. Journal of Vibration and Shock, 2016, 35(4): 127-133

参考文献

[1] 程军圣,张亢,杨宇,等. 局部均值分解与经验模式分解的对比研究[J]. 振动与冲击,2009,28(5):13-16.
Cheng Junsheng, Yang Yu, Yang Yu, et al. Comparative study between local mean decomposition and empirical mode decomposition [J]. Journal of Vibration and Shock, 2009, 28(5): 13-16.
[2] Wu Zhicheng, Wang Chongyang, Ren Aijun. Optimal selection of wavelet base functions for eliminating signal trend based on wavelet analysis[J]. Beijing Ligong Daxue Xuebao/Transaction of Beijing Institute of Technology, 2013, 33: 811-814.
[3] 罗洁思,于德介,彭富强,等.基于EMD的多尺度形态学解调方法及其在机械故障诊断中的应用[J].振动与冲击,2009,28(11):84-86.
Luo Jiesi, Yu Dejie, Peng Fuqiang, et al. Multi-scale morphological demodulation method based on EMD and its application in mechanical fault diagnosis [J]. Journal of Vibration and Shock, 2009, 28(11): 84-86.
[4] Liu Xiaofeng, Bo Lin, Luo Honglin. Bearing faults diagnostics based on hybrid LS-SVM and EMD method measurement [J]. Journal of the international measurement confederation, 2015, 59: 145-166.
[5] Wu Z H, Huang N E.Ensemble empirical mode decomposition: a noise-assisted data analysis method [J]. Advances in Adaptive Data Analysis, 2009, 1(1): 1-41.
[6] 丁康,李巍华,朱小勇. 齿轮及齿轮箱故障诊断实用技术[M].北京:机械工业出版社,2005:1−4.
DING Kang, LI Wei-hua, ZHU Xiao-yong. Practical technologyof gear and gear box fault diagnosis[M]. Beijing: China Machine Press, 2005: 1−4.
[7] 张亢,程军圣,杨宇. 基于自适应波形匹配延拓的局部均值分解端点效应处理方法[J].振动与冲击,2010,21(4):457-462.
    Zhang Kang, Cheng Jun-sheng, Yang Yu. Processing method for end effects of local mean decomposition based on self-adaptive waveform matching extending[J]. Journal of Vibration and Shock, 2010 ,21(4) : 457-462.
[8] Sweeney K T, McLoone S F 2013 IEEE Trans. Biomed. Eng. 60-97.
[9] Serra J.Morphological filtering:anoverwiew[J].Signal Process,1994,38(1):3-11.
[10] 李兵,张培林,米双山,等. 机械故障信号的数学形态学分析与智能分类.北京:国防工业出版社,2011.
   Li Bing, Zhang Pei-lin, MiShuang-shan, etal. Morphological analysis and intelligent classification Mechanical fault signal. Beijing:National Defense Industry Press, 2011.
[11] 章立军,杨德斌,徐金梧,等.基于数学形态滤波的齿轮故障特征提取方法[J].机械工程学报,2007,43(2):71-75
Zhang Lijun, Yang Debin, Xu Jinwu, et al. Approach to extracting gear fault feature based on mathematical morphological filtering [J] Chinese Journal of Mechanical Engineering, 2007, 43(2): 71-75.
[12] 胡爱军,孙敬敬,向玲,等.振动信号处理中数学形态滤波器频率响应特性研究 [J].机械工程学报,2012,48(1):99-103
Hu Aijun, Sun Jingjing, Xiang Ling. Analysis of Morphological Filter’s Frequency Response Characteristics in Vibration Signal Processing [J] Chinese Journal of Mechanical Engineering, 2012, 48(1): 99-103.
[13] Mandic D P. Filter Bank Property of Multivariate Empirical Mode Decomposition[J]. SignalProcessing, IEEE Transactions on, 2011, 59(5): 2421-2426.
[14] Harris M C, Blotter J D, Scott D. Sommerfeldt obtaining the complex pressure field at the hologram surface for use in near-field acoustical holography when pressure and in-plane velocities are measured [J]. The Journal of the Acoustical Society of America, 2006 119(2): 808-816.
[15] 赵昭,刘利林,张承学,等.形态学滤波器结构元素选取原则研究与分析[J].电力系统保护与控制,2009,37(14):21-25.
ZHAO Zhao,Liu li-lin,ZHANG Cheng-xue,etal.Large EEMD noise reduction method for non-stationary vibration signal of rotating machines [J].Power System Protection and Control,2009,37(14):21-25.
[16] Rehman N, Mandic D P. Multivariate Empirical Mode Decomposition[J]. Proceedings of the Royal Society A: Mathematical, Physical and Engineering Science, 2010, 466(2117):1291-1302.
[17] Rilling G, Flandrin P, Gonalves P, et al. Bivariate Empirical Mode Decomposition[J]. Signal Processing Letters, IEEE, 2007, 14(12): 936-939.
[18] Mandic D P. Empirical Mode Decomposition for Trivariate Signals[J]. Signal Processing, IEEE Transactions on, 2010, 58(3): 1059-1068.
[19] H.Niederreiter. Random Number Generation and Quasi-Monte Carlo Methods[M].Philadelphia,PA: Society for Industrial and Applied Mathematics, 1992: 1-158.
[20] Rilling G, Flandrin P, Gonçalves P 2003 Proceedings of IEEE-EURASIP Workshop on Nonlinear Signal andImage Processing Grado (I), June 2003 p1
[21] LEI Y G, HE Z J, and ZI Y Y, EEMD method and WNN for fault diagnosis of locomotive roller bearings [J]. Expert Systems with Applications, 2011, 38(6): 7334-7341.

PDF(2627 KB)

Accesses

Citation

Detail

段落导航
相关文章

/