基于改进拉普拉斯特征映射和约束种子K均值的半监督故障识别

PDF(1064 KB)

振动与冲击 ›› 2019, Vol. 38 ›› Issue (16) : 93-99.

论文

张鑫1,2，郭顺生1,2，江丽1,2

作者信息 +

Semi-supervised fault identification based on improved Laplace feature mapping and constraint seed K-means

ZHANG Xin1,2，GUO Shunsheng1,2，JIANG Li1,2

Author information +

文章历史 +

摘要

为充分利用少量有标记样本蕴含的重要信息，在拉普拉斯特征映射（LE）算法基础上，对标记样本点进行置信度约束，提出了改进的LE算法及基于该算法的半监督故障诊断模型。该模型采用改进的LE算法，直接从原始高维振动信号中提取最敏感的低维流形特征，随后将其输入到基于约束种子K均值算法构建的分类器，从而以可视化的聚类结果标识机械设备的运行状态。与核主成分分析、核判别分析等经典算法进行比较，该模型能明显提高轴承故障类型和滚动体故障严重性的识别性能。

Abstract

Aiming at making full use of the important messages contained in a small number of marked samples.The Laplacian eigenmap (LE) algorithm was improved by implementing confidence constraints on marked sample points.The semi-supervised fault diagnosis model based on the improved LE algorithm was presented.This model utilized the improved LE algorithm to extract the most sensitive low-dimensional manifold features from the raw high-dimensional vibration signals directly.Subsequently, they were fed into the classifier based on the constraint seed K-means algorithm.Thus, the operating conditions of mechanical equipment were identified by visual clustering results.Compared with the Kernel principal component analysis and the Kernel discriminant analysis, the model obviously improves the recognition performance of bearing fault types and ball fault severities.

导出引用

张鑫1,2，郭顺生1,2，江丽1,2. 基于改进拉普拉斯特征映射和约束种子K均值的半监督故障识别[J]. 振动与冲击, 2019, 38(16): 93-99

ZHANG Xin1,2，GUO Shunsheng1,2，JIANG Li1,2. Semi-supervised fault identification based on improved Laplace feature mapping and constraint seed K-means[J]. Journal of Vibration and Shock, 2019, 38(16): 93-99

参考文献

[1] Gao, Z., C. Cecati, and S.X. Ding, A Survey of Fault Diagnosis and Fault-Tolerant Techniques—Part I: Fault Diagnosis With Model-Based and Signal-Based Approaches[J]. IEEE Transactions on Industrial Electronics, 2015. 62(6): p. 3757-3767.
[2] Lei, Y., et al., An Intelligent Fault Diagnosis Method Using Unsupervised Feature Learning Towards Mechanical Big Data[J]. IEEE Transactions on Industrial Electronics, 2016. 63(5): p. 3137-3147.
[3] Pan, W., et al., Online fault diagnosis for nonlinear power systems ☆[J]. Automatica, 2015. 55: p. 27-36.
[4] 江丽，郭顺生, 基于半监督拉普拉斯特征映射的故障诊断[J]. 中国机械工程, 2016. 27(14): p. 1911-1916.
Jiang Li, Guo Shun-sheng. Fault Diagnosis Based on Semi-supervised Laplacian Eigenmaps[J]. China Mechanical Engineering, 2016. 27(14): p. 1911-1916.
[5] Ramahaleomiarantsoa J F, Heraud N, Sambatara E J R, et al. Principal components analysis method application in electrical machines diagnosis[C]// 8th International Conference on Informatics in Control, Automation and Robotics (ICINCO). 2011..
[6] Hyvärinen A, Oja E. Independent component analysis: algorithms and applications[J]. Neural Networks, 2000, 13(4):411-430.
[7] Zhi-Nong L I, Wang X Y, Zhang X G. Recognition Method of Rolling Bearing Fault Based on Kernel Principle Component Analysis[J]. Bearing, 2008.
[8] Hao T F, Guo C. Machinery fault diagnosis based on Bayes optimal kernel discriminant analysis[J]. Journal of Vibration & Shock, 2012, 31(13):26-30.
[9] Owsley L M D, Atlas L E, Bernard G D. Self-organizing feature maps and hidden Markov models for machine-tool monitoring[J]. Signal Processing IEEE Transactions on, 1997, 45(11):2787-2798.
[10] Cayton L. Algorithms for manifold learning[J]. Univ.of California at San Diego Tech.rep, 2005.
[11] Belkin M, Niyogi P. Laplacian Eigenmaps for Dimensionality Reduction and Data Representation[J]. Neural Computation, 2006, 15(6):1373-1396.
[12] 黄宏臣, 韩振南, 张倩倩,等. 基于拉普拉斯特征映射的滚动轴承故障识别[J]. 振动与冲击, 2015, 34(5):128-134.
Huang Hong-chen, Han Zhen-nan, Zhang Qian-qian, et al. Method of Fault Diagnosis for Rolling Bearings Based on Laplacian Eigenmap[J].Journal of Vibration and Shock,2015，34(5):128-134.
[13] 蒋全胜, 贾民平, 胡建中,等. 基于拉普拉斯特征映射的故障模式识别方法[J]. 系统仿真学报, 2008(20):5710-5713.
Jiang Quan-sheng, Jia Min-ping, Hu Jian-zhong, et al. Method of Fault Pattern Recognition Based on Laplacian Eigenmaps[J]. Journal of System SImulation, 2008(20):5710-5713.
[14] 李月仙, 韩振南, 黄宏臣,等. 基于拉普拉斯特征映射的旋转机械故障识别[J]. 振动与冲击, 2014, 33(18):21-25.
Li Yue-xian, Han Zhen-nan, Huang Hong-chen,et al. Fault diagnosis of rotating machineries based on Laplacian eigenmaps[J].Journal of Vibration and Shock, 2014, 33(18):21-25.
[15] Ng M K. A note on constrained k -means algorithms[J]. Pattern Recognition, 2000, 33(3):515-519..
[16] 王泽杰, 胡浩民. 流形学习算法中的参数选择问题研究[J]. 计算机应用与软件, 2010, 27(6):84-85..
Wang Ze-jie, Hu Hao-min. On Parameter Selection In Manifold Learning Algorithm[J].Computer Applications and Software, 2010, 27(6):84-85.
[17] 李昆仑, 曹铮, 曹丽苹,等. 半监督聚类的若干新进展[J]. 模式识别与人工智能, 2009, 22(5):735-742.
Li Kun-lun, Cao Zheng, Cao Li-ping,et al. Some Developments on Semi-Supervised Clustering[J].PR&AI, 2009, 22(5):735-742.
[18] 张永晶. 初始聚类中心优化的K-means改进算法[D]. 东北师范大学, 2013.
Zhang Yong-jing. Improved K-Means Algorithm Based on Optimizing Initial Cluster Centers[D]. Northeast Normal University,2013.
[19] Loparo K A. Bearings Vibration Data Set[DB/OL]. Cleveland, Ohio:Case Western Reserve University. [2014-10-28]. http://Csegroups. Case. Edu/Bearingdatacenter/Home.
[20] 李刚, 邢书宝, 薛惠锋. 基于RBF核的SVM及RVM模式
分析性能比较[J]. 计算机应用研究, 2009, 26(5):1782-1784.
Li Gang, Xing Shu-bao, Xue Hui-feng. Comparison on Pattern Analysis Performance of SVM and RVM Based on RBF Kemel[J]. Application Research of Computers, 2009, 26(5):1782-1784.
[21] 赵孝礼, 赵荣珍. 全局与局部判别信息融合的转子故障数据集降维方法研究[J]. 自动化学报, 2017, 43(4):560-567.
Zhao Xiao-li, Zhao Rong-zhen. A Method of DImension Reduction of Rotor Faults Data Set Based on Fusion of Global and Local Discriminant Information[J]. ACTA AUTOMATICA SINICA, 2017, 43(4):560-567.