基于特征处理的MVU算法在齿轮故障诊断中的应用

摘要
图/表
参考文献(17)
相关文章 (15)

全文: PDF (2137 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要针对不同故障类别齿轮的故障信息难以有效获取、齿面多类故障难以准确聚类的问题，提出一种基于特征处理的最大方差展开（Maximum Variance Unfolding，MVU）维数简约的齿轮故障诊断模型。首先对获取的振动信号进行最小熵反卷积（Minimum Entropy Deconvolution，MED）预处理，将高低频段进行分离并筛除不确定信号，并在多域上提取信息熵作为特征指标；而后，利用样本点分布矩阵筛选高效表征特征指标并构建高维特征空间，并利用改进的MVU算法对其进行维数简约，获取低维的真实子空间；最后，将其输入到超球多类支持向量机中进行超球构造与分类识别。通过实验数据的分析对比验证模型的有效性。

	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
	陈俊康
	陈小虎
	王旭平
	蒋成伟

关键词 ：信息熵, 特征处理, 最大方差展开, 超球多类支持向量机, 齿轮

Abstract：Aiming at problems of gear fault information being difficult to effectively extract and gear surface faults being difficult to correctly cluster, a gear fault diagnosis model using the maximum variance unfolding (MVU) algorithm based on feature processing was proposed here. Firstly, the collected faulty gear vibration signals were pre-processed with the minimum entropy deconvolution (MED), higher and lower frequency bands were separated and uncertain signals were excluded. The information entropy was extracted in multi-domain as feature indexes. Then efficient feature indexes were screened out with the sample point distribution matrix and they were used to construct high dimensional feature space, and the improved MVU algorithm was used to do dimension reduction, and obtain an actual subspace with lower dimension. Finally, the actual subspace was input into a hyper-sphere multi-class SVM to do hyper-sphere construction and classification identification. The effectiveness of the proposed model was verified through contrastively analyzing test data.

Key words： information entropy feature processing maximum variance unfolding hyper-sphere multiclass support vector machine gear

收稿日期: 2018-06-22 出版日期: 2019-12-28

引用本文:

陈俊康，陈小虎，王旭平，蒋成伟. 基于特征处理的MVU算法在齿轮故障诊断中的应用[J]. 振动与冲击, 2020, 39(1): 123-130.
CHEN Junkang，CHEN Xiaohu，WANG Xuping，JIANG Chengwei. Application of MVU algorithm based on feature processing in gear fault diagnosis. JOURNAL OF VIBRATION AND SHOCK, 2020, 39(1): 123-130.

链接本文:

http://jvs.sjtu.edu.cn/CN/ 或 http://jvs.sjtu.edu.cn/CN/Y2020/V39/I1/123

[1] 姜景升, 王华庆, 柯燕亮, 向伟. 基于LTSA与K-最近邻分类器的故障诊断[J]. 振动与冲击, 2017, 36(11):134-139.
JIANG Jing-sheng, WANG Hua-qing, KE Yan-liang. Fault Diagnosis Based on LTSA and K-Nearest Neighbor Classifier [J]. Journal of Vibration and Shock, 2017, 36(11):134-139.
[2] 宋涛, 汤宝平, 李锋. 基于流形学习和K-最近邻分类器的旋转机械故障诊断方法[J]. 振动与冲击, 2013, 32(5):149-153.
SONG Tao, TANG Bao-ping, LI Feng. Fault diagnosis method for rotating machinery based on manifold learning and K-nearest neighbor classifier [J]. Journal of Vibration and Shock, 2013, 32(5):149-153.
[3] 张晓涛, 唐力伟, 王平, 邓士杰. 基于多尺度正交PCA-LPP流形学习算法的故障特征增强方法[J]. 振动与冲击, 2015, 34(13):66-70.
ZHANG Xiao-tao, TANG Li-wei, WANG Ping, DENG Shi-jie. Fault feature enhancement method based on multiscale orthogonal PCA-LPP manifold learning algorithm [J]. Journal of Vibration and Shock, 2015, 34(13):66-70.
[4] Weinberger K Q, Sha F, Saul L K. Learning a kernel matrix for nonlinear dimensionality reduction[C]// International Conference on Machine Learning. ACM, 2004:106.
[5] 朱美琳, 刘向东, 陈世福. 用球结构的支持向量机解决多分类问题[J]. 南京大学学报(自然科学), 2003, 39(2):153-158.
ZHU Mei-lin, LIU Xiang-dong, CHEN Shi-fu. Solving the Problem of Multi-class Pattern Recognition with Sphere-structured Support Vector Machines [J]. Journal of Nanjing University (Natural Sciences), 2003, 39(2):153-158.
[6] Wiggins R A. Minimum entropy deconvolution [J]. Geophysical Prospecting for Petrole, 1980, 16(1):21-35.
[7] Nandi A K, Mampel D, Roscher B. Blind deconvolution of ultrasonic signals in nondestructive testing applications [J]. Signal Processing IEEE Transactions on, 1997, 45(5):1382-1390.
[8] Pincus S M. Assessing serial irregularity and its implications for health [J]. Annals of the New York Academy of Sciences, 2001, 954(1):245.
[9] Cui B Z, Pan H X, Wang Z B. Fault diagnosis of roller bearings base on the local wave and approximate entropy [J]. Journal of North University of China, 2012, 33(5):552-558.
[10] 何志坚, 周志雄. 基于ELMD的样本熵及Boosting-SVM的滚动轴承故障诊断[J]. 振动与冲击, 2016, 35(18):190-195.
HE Zhi-jian, ZHOU Zhi-xiong. Fault diagnosis of roller bearings based on ELMD sample entropy and Boosting-SVM [J]. Journal of Vibration and Shock, 2016, 35(18):190-195.
[11] Zhu K. Fault Diagnosis of Rolling Bearings Based on IMF Envelope Sample Entropy and Support Vector Machine [J]. Journal of Information & Computational Science, 2013, 10(16):5189-5198.
[12] Borchers B. CSDP, AC library for semidefinite programming [J]. Optimization Methods & Software, 1999, 11(1-4):613-623.
[13] Lkopf B, Smola A J, Williamson R C, et al. New Support Vector Algorithms [J]. Neural Computation, 2000, 12(5):1207-1245.
[14] 赵孝礼, 赵荣珍, 孙业北,等. 基于正则化核最大边界投影维数约简的滚动轴承故障诊断[J]. 振动与冲击, 2017, 36(14):104-110.
ZHAO Xiao-li, ZHAO Rong-zhen, SUN Ye-bei, et al. Fault diagnosis of rolling bearings based on the dimension reduction using the regularized kernel maximum margin projection [J]. Journal of Vibration and Shock, 2017, 36(14):104-110.
[15] 陈如清. 基于KPCA-MVU的噪声非线性过程故障检测方法[J]. 仪器仪表学报, 2014(12):2673-2680.
CHEN Ru-qing. Nonlinear process fault detection method under noise environment using KPCA and MVU [J]. Chinese Journal of Scientific Instrument, 2014(12):2673-2680.
[16] 康守强, 王玉静, 姜义成,等. 基于超球球心间距多类支持向量机的滚动轴承故障分类[J]. 中国电机工程学报, 2014, 34(14):2319-2325.
KANG Shou-qiang, WANG Yu-jing, JIANG Yi-cheng, et al. Fault Classification of Rolling Bearing Based on Hypersphere Spherical Center Distance Multiclass Support Vector Machine [J]. Proceeding of the CSEE, 2014, 34(14):2319-2325.
[17] 柴凯, 张梅军, 黄杰,等. 基于时频特征和PCA-KELM的液压系统故障诊断[J]. 解放军理工大学学报(自然科学版), 2015(4):394-400.
CHAI Kai, ZHANG Meijun, HUANG Jie, et al. Fault diagnosis of hydraulic system based on time- frequency characteristics and PCA-KELM [J]. Journal of PLA University of Science and Technology (Natural Science Edition), 2015(4):394-400.