相关流形距离在转子故障数据集分类中的应用方法
|
摘要
针对故障特征属性值域之间存在着一定相关性导致准确分类困难的问题,提出一种能够考虑相关系数影响作用的转子故障数据集分类方法;该方法是将相关流形距离的边界Fisher分析(Correlation Manifold Distance Marginal Fisher Analysis,CDMFA)与相关流形距离的K-近邻(Correlation Manifold Distance K-Nearest Neighbor,CDKNN)分类器概念相结合在一起的结果。首先,将振动信号集合转换成多域、多通道高维故障特征数据集;然后,通过CDMFA将融合相关系数的相关流形距离用于度量数据样本间的近邻与权值,据此能更好地反映高维数据间的相似性关系,提取出能使类间距离趋大的低维特征子集;最后,将得到的低维特征子集输入到CDKNN分类器中进行故障模式辨识。用一个双跨度转子系统数据集与仿真数据集对所提出的方法进行了验证。结果表明:本方法降维效果良好,可获得更高的故障分类准确率。研究发现,采用相关流形距离作为信息测度的故障数据分类方法能更真实地揭示出高维特征间的几何结构关系;该方法可为高维故障数据集的特征属性约简与分类,提供降低数据规模的理论参考依据。
Aiming at the difficulty in the accurate classification of rotor fault data due to a certain correlation between fault feature attribute domains,a kind of rotor fault data classification method considering the influence of the correlation coefficient was proposed. The method was based on the concept combining the correlation manifold distance marginal Fisher analysis( CDMFA) and correlation manifold distance K-nearest neighbor( CDKNN) classifier together.First of all,vibration signals were converted into high-dimensional data-set in multi-domain and multi-channel. Then,using the correlation manifold distance of the fused correlation coefficient to measure the neighbors and weights of fault samples by the CDMFA,which can better reflect the similarity relation between high-dimensional data and extract lowdimensional feature subsets to make the distance bigger between the classes. Finally,the low-dimensional feature subsets were input into the CDKNN classifier for fault pattern recognition. The proposed method was verified by using a double span rotor system data-set and simulation data-set. The results show that the method has better dimension reduction effect and higher fault classification accuracy. The study finds that the manifold distance fault data classification method can reveal more realistic geometrical relation between high-dimensional features. The method provides a theoretical reference to the data preprocessing for feature attribute reduction and classification of high dimensional fault data-set.
Aiming at the difficulty in the accurate classification of rotor fault data due to a certain correlation between fault feature attribute domains,a kind of rotor fault data classification method considering the influence of the correlation coefficient was proposed. The method was based on the concept combining the correlation manifold distance marginal Fisher analysis( CDMFA) and correlation manifold distance K-nearest neighbor( CDKNN) classifier together.First of all,vibration signals were converted into high-dimensional data-set in multi-domain and multi-channel. Then,using the correlation manifold distance of the fused correlation coefficient to measure the neighbors and weights of fault samples by the CDMFA,which can better reflect the similarity relation between high-dimensional data and extract lowdimensional feature subsets to make the distance bigger between the classes. Finally,the low-dimensional feature subsets were input into the CDKNN classifier for fault pattern recognition. The proposed method was verified by using a double span rotor system data-set and simulation data-set. The results show that the method has better dimension reduction effect and higher fault classification accuracy. The study finds that the manifold distance fault data classification method can reveal more realistic geometrical relation between high-dimensional features. The method provides a theoretical reference to the data preprocessing for feature attribute reduction and classification of high dimensional fault data-set.
引文
[1]LEI Y,HE Z,ZI Y.Application of the EEMD method to rotor fault diagnosis of rotating machinery[J].Mechanical Systems&Signal Processing,2009,23(4):1327-1338.
[2]艾延廷,费成巍,王志.基于改进模糊SVM的转子振动故障诊断技术[J].航空动力学报,2011,26(5):1118-1123.AI Yanting,FEI Chengwei,WANG Zhi.Technique for diagnosing fault of rotor vibration based on improved fuzzy SVM[J].Journal of Aerospace Power,2011,26(5):1118-1123.
[3]LEI Y,HE Z,ZI Y.A new approach to intelligent fault diagnosis of rotating machinery[J].Expert Systems with Applications,2008,35(4):1593-1600.
[4]SU Z,TANG B,MA J,et al.Fault diagnosis method based on incremental enhanced supervised locally linear embedding and adaptive nearest neighbor classifier[J].Measurement,2014,48(1):136-148.
[5]JIANG Q,JIA M,HU J,et al.Machinery fault diagnosis using supervised manifold learning[J].Mechanical Systems&Signal Processing,2009,23(7):2301-2311.
[6]YAN S,XU D,ZHANG B,et al.Graph embedding and extension:A general framework for dimensionality reduction[J].Pattern Analysis&Machine Intelligence IEEE Transactions on,2007,29(1):40-51.
[7]JIANG L,SHI T,XUAN J.Fault diagnosis of rolling bearings based on marginal Fisher analysis[J].Journal of Vibration&Control,2014,20(3):470-480.
[8]XU D,YAN S,TAO D,et al.Marginal Fisher analysis and its variants for human gait recognition and content-based image retrieval[J].IEEE Transactions on Image Processing,2007,16(11):2811-2821.
[9]史骏,陈才扣.基于马氏距离的半监督鉴别分析及人脸识别[J].北京航空航天大学学报,2011,37(12):1589-1593.SHI Jun,CHEN Caikou.Mahalanobis distance-based semisupervised discriminant analysis for face recognition[J].Journal of Beijing University of Aeronautics and Astronautics,2011,37(12):1589-1593.
[10]刘嘉敏,罗甫林,黄鸿,等.应用相关近邻局部线性嵌入算法的高光谱遥感影像分类[J].光学精密工程,2014,22(6):1668-1676.LIU Jiamin,LUO Fulin,HUANG Hong,et al.Classification of hyperspectral remote sensing images using correlation neighbor LLE[J].Optics and Precision Engineering,2014,22(6):1668-1676.
[11]FALIE D,DAVID L.Correlation coefficient based on independent component analysis[C]//International Conference on Communications,2012:59-62.
[12]BLAKE C L,MERZ C J,1998,UCI repository of machine learning databases[EB/OL].[2016-01-20].University of California.Available online at:http://archive.ics.uci.edu/ml/.
[13]霍天龙,赵荣珍,胡宝权.基于熵带法与PSO优化的SVM转子故障诊断[J].振动、测试与诊断,2011,31(3):279-284.HUO Tianlong,ZHAO Rongzhen,HU Baoquan.Fault diagnosis for rotor systems based on entropy band method and support vector machine optimized by PSO[J].Journal of Vibration,Measurement&Diagnosis,2011,31(3):279-284.
[14]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 Zejie,HU Haomin.On parameter selection in manifold learning algorithm[J].Computer Applications and Software,2010,27(6):84-85.
[15]CHEN F,TANG B,CHEN R.A novel fault diagnosis model for gearbox based on wavelet support vector machine with immune genetic algorithm[J].Measurement,2013,46(1):220-232.
[2]艾延廷,费成巍,王志.基于改进模糊SVM的转子振动故障诊断技术[J].航空动力学报,2011,26(5):1118-1123.AI Yanting,FEI Chengwei,WANG Zhi.Technique for diagnosing fault of rotor vibration based on improved fuzzy SVM[J].Journal of Aerospace Power,2011,26(5):1118-1123.
[3]LEI Y,HE Z,ZI Y.A new approach to intelligent fault diagnosis of rotating machinery[J].Expert Systems with Applications,2008,35(4):1593-1600.
[4]SU Z,TANG B,MA J,et al.Fault diagnosis method based on incremental enhanced supervised locally linear embedding and adaptive nearest neighbor classifier[J].Measurement,2014,48(1):136-148.
[5]JIANG Q,JIA M,HU J,et al.Machinery fault diagnosis using supervised manifold learning[J].Mechanical Systems&Signal Processing,2009,23(7):2301-2311.
[6]YAN S,XU D,ZHANG B,et al.Graph embedding and extension:A general framework for dimensionality reduction[J].Pattern Analysis&Machine Intelligence IEEE Transactions on,2007,29(1):40-51.
[7]JIANG L,SHI T,XUAN J.Fault diagnosis of rolling bearings based on marginal Fisher analysis[J].Journal of Vibration&Control,2014,20(3):470-480.
[8]XU D,YAN S,TAO D,et al.Marginal Fisher analysis and its variants for human gait recognition and content-based image retrieval[J].IEEE Transactions on Image Processing,2007,16(11):2811-2821.
[9]史骏,陈才扣.基于马氏距离的半监督鉴别分析及人脸识别[J].北京航空航天大学学报,2011,37(12):1589-1593.SHI Jun,CHEN Caikou.Mahalanobis distance-based semisupervised discriminant analysis for face recognition[J].Journal of Beijing University of Aeronautics and Astronautics,2011,37(12):1589-1593.
[10]刘嘉敏,罗甫林,黄鸿,等.应用相关近邻局部线性嵌入算法的高光谱遥感影像分类[J].光学精密工程,2014,22(6):1668-1676.LIU Jiamin,LUO Fulin,HUANG Hong,et al.Classification of hyperspectral remote sensing images using correlation neighbor LLE[J].Optics and Precision Engineering,2014,22(6):1668-1676.
[11]FALIE D,DAVID L.Correlation coefficient based on independent component analysis[C]//International Conference on Communications,2012:59-62.
[12]BLAKE C L,MERZ C J,1998,UCI repository of machine learning databases[EB/OL].[2016-01-20].University of California.Available online at:http://archive.ics.uci.edu/ml/.
[13]霍天龙,赵荣珍,胡宝权.基于熵带法与PSO优化的SVM转子故障诊断[J].振动、测试与诊断,2011,31(3):279-284.HUO Tianlong,ZHAO Rongzhen,HU Baoquan.Fault diagnosis for rotor systems based on entropy band method and support vector machine optimized by PSO[J].Journal of Vibration,Measurement&Diagnosis,2011,31(3):279-284.
[14]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 Zejie,HU Haomin.On parameter selection in manifold learning algorithm[J].Computer Applications and Software,2010,27(6):84-85.
[15]CHEN F,TANG B,CHEN R.A novel fault diagnosis model for gearbox based on wavelet support vector machine with immune genetic algorithm[J].Measurement,2013,46(1):220-232.