基于复杂网络优化的DAG-SVM在滚动轴承故障诊断中的应用

摘要
图/表
参考文献(17)
相关文章 (15)

全文: PDF (1648 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要针对滚动轴承故障与其演化程度组合类型数量大，一般模式识别方法难以适应的问题，提出基于复杂网络优化的有向无环图支持向量机（CNDAG-SVM）。该方法引入复杂网络理论中相似性测度概念用以评定各样本类型间的分离性质，并以平均相似性测度作为有效度量样本类型可区分程度的测度对有向无环图叶节点类型进行排序，依次提取对应二元分类器构造较优有向无环图拓扑结构，缓解误差累积效应的同时提高了结构上层节点的容错能力，获得较高的正确识别率。利用局部均值分解方法提取乘积函数(Production Function，PF)分量波峰系数、峭度系数及能量构造特征向量，将其输入CNDAG-SVM分类器中用于区分滚动轴承的故障类型与演化程度。对滚动轴承内圈故障、外圈故障及滚动体故障振动信号的分析结果表明，该方法能准确有效识别故障类型与其演化程度，较之传统多元分类支持向量机具有更高的识别精度和效率。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章

关键词 ：复杂网络, 有向无环图支持向量机, 滚动轴承, 故障诊断

Abstract：

Due to the large quantities of crossed combinations of the fault patterns and evolution stages of rolling bearings, the general patterns recognition method was difficult to adapt to multivariate process. In view of the problem, an optimized directed acyclic graph support vector machine based on complex network was proposed. According to the similarity measure in complex network theory, the separate characters of samples were evaluated, and the nodes of directed acyclic graph were sequenced by the average similarity measure which was calculated as the criterion of distinguished degree of samples. Then the corresponding binary support vector machines were selected to construct the optimal directed acyclic graph, which had achieved high correction identification ratio by alleviating error accumulation and improving fault tolerance of the upper nodes. Feature vectors were constructed from the crest factor, kurtosis coefficient and energy of product functions which were obtained by local mean decomposition. And then the feature vectors were served as input parameters of CNDAG-SVM classifier to sort fault patterns and evolution stages of rolling bearings. By analyzing the vibration signal acquired from the bearings with inner-race, outer-race or elements faults, the experimental results indicate that the proposed method could recognize the types and evolution grades effectively and has higher accurateness and productiveness than traditional multi-class support vector machines.

Key words： complex network DAG-SVM rolling bearing fault diagnosis

收稿日期: 2014-03-31 出版日期: 2015-06-25

引用本文:

石瑞敏，杨兆建. 基于复杂网络优化的DAG-SVM在滚动轴承故障诊断中的应用[J]. 振动与冲击, 2015, 34(12): 1-6.
SHI Rui-min YANG Zhao-jian. Application of Optimized Directed Acyclic Graph support vector machine based on complex network in fault diagnosis of rolling bearing. JOURNAL OF VIBRATION AND SHOCK, 2015, 34(12): 1-6.

链接本文:

http://jvs.sjtu.edu.cn/CN/ 或 http://jvs.sjtu.edu.cn/CN/Y2015/V34/I12/1

[1] 钟秉林，黄仁. 机械故障诊断学[M]. 北京：机械工业出版社，2007:1-16.
ZHONG Bing-lin, HUANG Ren. Introduction to machine fault diagnosis[M]. Beijing: China Machine Press, 2007:1-16.
[2] 姚培，王仲生，姜洪开等. 局部保形映射和AdaBoost方法在滚动轴承故障诊断中的应用[J]. 振动与冲击，2013,32（5）：144-148.
YAO Pei, WANG Zhong-sheng, JIANG Hong-kai, et al. Roller bearing fault diagnosis based on locality preserving project [J]. Journal of Vibration and Shock, 2013, 32(5): 144-148.
[3] Candès E J, Charlton P R, Helgason H. Detecting highly oscillatory signal by chirplet path pursuit[J]. Applied and Computational Harmonic Analysis, 2008, 24(1): 14-40.
[4] 姜锐红，刘树林，刘颖慧等. 基于CPWP混合原子分解的滚动轴承故障诊断方法研究[J]. 振动与冲击，2013,32（23）：48-51.
JIANG Rui-hong, LIU Shu-lin, LIU Ying-hui, et al. Rolling bearing fault diagnosis based on CPWP merged atomic decomposition [J]. Journal of Vibration and Shock, 2013, 32(23): 48-51.
[5] Ocak H, Loparo K A, Discenzo F M. Online tracking of bearing wear using wavelet packet decomposition and probabilistic modeling: a method for bearing prognostics [J]. Journal of Sound and Vibration, 2007, 302(4/5): 951-961.
[6] 罗颂荣，程军圣，杨宇. 基于本征时间尺度分解和变量预测模型模式识别的机械故障诊断[J]. 振动与冲击，2013,32（13）：43-48.
LUO Song-rong, CHENG Jun-sheng, YANG Yu. Machine fault diagnosis method using ITD and variable predictive model-based class discrimination [J]. Journal of Vibration and Shock, 2013, 32(13): 43-48.
[7] 胥永刚，孟志鹏，陆明. 基于双树复小波包变换和SVM的滚动轴承故障诊断方法[J]. 航空动力学报，2014，29（1）：67-73.
XU Yong-gang, MENG Zhi-peng, LU Ming. Fault diagnosis method of rolling bearing based on dual-tree complex wavelet packet transform and SVM [J]. Journal of Aerospace Power, 2014, 29(1): 67-73.
[8] Vapnik V N. Statistical learning theory[M]. New York: Springer, 1998.
[9] 朱波，刘飞，李顺江.基于优化有向无环图支持向量机的多变量过程均值异常识别[J]. 计算机集成制造系统，2013,19（3）：549-568.
ZHU Bo, LIU Fei, LI Shun-jiang. Mean abnormality identification in multivariate process based on optimized directed acyclic graph support vector machine [J]. Computer Integrated Manufacturing Systems, 2013, 19(3): 549-568.
[10] 陈世杰，连可，王厚军. 遗传算法优化的SVM模拟电路故障诊断方法[J]. 电机科技大学学报，2009,38（4）：553-558.
CHEN Shi-jie, LIAN Ke, WANG Hou-jun. Method for analog circuit fault diagnosis based on GA optimized SVM [J]. Journal of University of Electronic Science and Technology of China, 2009, 38(4): 553-558.
[11] 陈安华，潘阳，蒋玲莉. 基于复杂网络社团聚类的故障模式识别方法研究[J]. 振动与冲击，2013，32（20）：129-133.
CHEN An-hua, PAN Yang, JIANG Ling-li. Fault pattern recognition method based on complex network community clustering [J]. Journal of Vibration and Shock, 2013, 32(20): 129-133.
[12] 杜海峰，王娜，张进花等. 基于复杂网络的故障诊断策略[J]. 机械工程学报，2010,46（3）：90-96.
DU Hai-feng, WANG Na, ZHANG Jin-hua, et al. Fault diagnosis strategy based on complex network analysis [J]. Journal of Mechanical Engineering, 2010, 46(3): 90-96.
[13] Newman M E J. Detecting community structure in network [J]. Eur. Phys. J. B, 2004, 38(2):321-330.
[14] 刘卫兵，李志农，蒋静. 基于局域均值分解的滚动轴承故障诊断方法[J]. 轴承，2009（9）：48-52.
LIU Weibing, LI Zhinong, JIANG Jing. Fault diagnosis method of rolling bearing based on local mean decomposition [J]. Bearing, 2009(9): 48-52.
[15] Smith J S. The local mean decomposition and its application to EEG perception data [J]. Journal of the Royal Society Interface, 2005, 2(5): 443-454.
[16] 唐贵基，王晓龙. 基于局部均值分解和切片双谱的滚动轴承故障诊断研究[J]. 振动与冲击，2013，32（24）：83-88.
TANG Gui-ji, WANG Xiao-long. Fault diagnosis of roller bearing based on local mean decomposition and slice bispectrum [J]. Journal of Vibration and Shock, 2013, 32(24): 83-88.
[17] http://www.eecs.cwru.edu/laboratory/bearing, Bearing Data Center Website, Case Western Reserve University