基于LTSA与K-最近邻分类器的故障诊断

姜景升,王华庆,柯燕亮,向伟

振动与冲击 ›› 2017, Vol. 36 ›› Issue (11) : 134-139.

PDF(901 KB)
PDF(901 KB)
振动与冲击 ›› 2017, Vol. 36 ›› Issue (11) : 134-139.
论文

基于LTSA与K-最近邻分类器的故障诊断

  • 姜景升,王华庆,柯燕亮,向伟
作者信息 +

Fault Diagnosis Methods Based on Local Tangent Space Alignment and K-Nearest Neighbor Classifier

  • Jingsheng Jiang, Huaqing Wang, Yanliang Ke, Wei Xiang
Author information +
文章历史 +

摘要

针对局部切空间排列算法(LTSA)的效果受近邻数k值影响较大的缺点,提出基于聚类准则的LTSA与K-最近邻分类器的故障诊断模型。首先基于振动信号的时域特征构建高维特征矩阵;然后对高维矩阵进行标准化预处理,依据聚类准则确定局部切空间排列中的最佳近邻数k,运用LTSA提取高维矩阵的低维特征向量;最后将提取的低维特征向量利用K-最近邻分类器进行故障模式识别。采用轴承诊断实验系统进行验证,结果表明,基于聚类准则的优化方法可有效地克服近邻数k选择的盲目性,提高了局部切空间的降维精度和故障模式识别正确率,其在轴承时域特征维数约简方面,效果优于主成分分析(PCA)与拉普拉斯特征映射(LE),适用于轴承故障诊断。

Abstract

Aiming at the problem that the performance of local tangent space alignment(LTSA) is greatly influenced by nearest neighbor k, the fault diagnosis mode of LTSA and K-nearest neighbor classifier(KNN)based on clustering criterion is proposed. First, the vibration signal collected was used to construct high dimensional matrix, then the high dimensional matrix was standardized before dimensionality reduction. According to clustering criterion, the nearest neighbor k in LTSA is chosen, and then the low dimensional feature vector of high dimensional matrix in the LTSA was extracted. Finally, the extracted low dimensional feature vectors were put into KNN to do fault pattern recognition. The results from fault diagnosis experimental system show that this method based on clustering criterion can effectively overcome the choice blindness of the optimal neighbor number k, and improve the precision of dimension reduction and recognition rate of the fault diagnosis pattern. Compared with the principal component analysis (PCA) method and Laplacian eigenmaps(LE)method, the method proposed is more suitable for bearing fault pattern recognition.

关键词

局部切空间排列 / K-最近邻分类器 / 聚类准则 / 故障诊断

Key words

local tangent space alignment (LTSA) / K-nearest neighbor (KNN) / clustering criterion / fault diagnosi

引用本文

导出引用
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
Jingsheng Jiang, Huaqing Wang, Yanliang Ke, Wei Xiang. Fault Diagnosis Methods Based on Local Tangent Space Alignment and K-Nearest Neighbor Classifier[J]. Journal of Vibration and Shock, 2017, 36(11): 134-139

参考文献

[1] 刘丽娟, 陈果, 郝腾飞. 基于流行学习与一类支持向量机的滚动轴承早期故障识别方法[J]. 中国机械工程, 2013, 24(5): 629-633.
Liu Lijuan, Chen Guo, Hao Tengfei. Incipient fault recognition of rolling bearings based on manifold learning and one–class SVM[J]. China Mechanical Engineering, 2013, 24(5): 629-633.
[2] 李金荣, 王国英, 莫路锋. 基于感知数据时域特征的WSNs故障被动诊断方法[J]. 传感技术学报, 2015, 28(7): 1078-1085.
Li Jinrong, Wang Guoying, Mo Lufeng. Passive diagnosis for WSNs using time domain features of sensing data[J]. Journal of Transduction Technology, 2015, 28(7): 1078-1085.
[3] 陈珍, 夏靖波, 柏骏. 基于进化深度学习的特征提取算法[J]. 计算机科学, 2015, 42(11): 288-292.
Chen Zhen, Xia Jingbo, Bo jun. Feature extraction algorithm based on evolutionary deep learning[J]. Computer Science, 2015, 42(11): 288-292.
[4] RABLT, SADOGHIM, JACOBSENHA. Solving big data challenges for enterprise application performance management[J]. Proceedings of the VLDB Endowment, 2012, 5(12): 1724-1735.
[5] LaSalle, Dominique, Karypis, George. MPI for Big Data: New tricks for an old dog[J]. Parallel Computing, 2014, 40(10): 754-767.
[6] Seung H S, Daniel D L. The Manifold Ways of Perception[J]. Science, 2000, 290(5500): 2268-2269.
[7] Roweis S, Saul L. Nonlinear Dimensionality Reduction by Locally Linear Embedding[J]. Science, 2000, 290(5500): 2323-2326.
[8] Tenenbaum J, Silva DD, Langford J. A Global Geometric Framework for Nonlinear Dimensionality Reduction[J]. Science, 2000, 290(5500): 2319-2323.
[9] 万鹏, 王红军, 徐小力. 局部切空间排列和支持向量机的故障诊断模型[J]. 仪器仪表学报, 2012, 33(12): 2790-2795.
Wan Peng, Wang Hongjun, Xu Xiaoli. Fault diagnosis model based on local tangent space alignment and support vector machine[J]. Chinese Journal of Scientific Instrument, 2012, 33(12): 2790-2795.
[10] 李锋, 汤宝平, 董绍江. 基于正交邻域保持嵌入特征约简的故障诊断模型[J]. 仪器仪表学报, 2011, 32(3): 622-627.
Li Feng, Tang Baoping, Dong Shaojiang. Fault diagnosis model based on feature compression with orthogonal neighborhood preserving embedding[J]. Chinese Journal of Scientific Instrument, 2011, 32(3): 622-627.
[11] 郑伟勇, 李艳玮. 降维方法在人脸图像识别中的分析与评估[J]. 工业控制计算机, 2015, 28(7): 110-112.
Zheng Weiyong, Li Yanwei. Dimension reduction method in face image recognition system. Industrial Control Computer, 2015, 28(7): 110-112.
[12] 王洁, 杨平, 郁嵩. 主成分分析和小波神经网络在气缸疲劳失效预测中的应用[J]. 机床与液压, 2015, 43(13): 167-171.
Wang Jie, Yang Ping, Yu Song. Application of principal component analysis and wavelet neural network for prediction of cylinder fatigue failure. Machine Tool & Hydraulics, 2015, 43(13): 167-171.
[13] 陈法法, 汤宝平, 苏祖强. 基于等距映射与加权KNN的旋转机械故障诊断[J]. 仪器仪表学报, 2013, 34(1): 216-220.
Chen Fafa, Tang Baoping, Su Zuqiang. Rotating machinery fault diagnosis based on isometric mapping and weighted KNN[J]. Chinese Journal of Scientific Instrument, 2013, 34(1): 216-220.
[14] 李胜, 张培林,吴定海. 基于渐近权值小波降噪和算法的液压泵Adaboost故障诊断[J]. 中国机械工程, 2011, 22(9): 1067-1071.
Li Sheng, Zhang Peilin, Wu Dinghai. Fault diagnosis for hydraulic pump based on gradual asymptotic weight selection of wavelet and adaboost. China Mechanical Engineering, 2011, 22(9): 1067-1071.
[15] 于德介, 陈淼峰, 程军圣. 一种基于支持向量机预测器模型的转子系统故障诊断方法[J]. 中国机械工程, 2006, 17(7): 696-699.
Yu Dejie, Chen Miaofeng, Cheng Junsheng. Fault diagnosis approach for rotor systems based on support vector machine predictive model[J]. China Mechanical Engineering, 2006, 17(7): 696-699.
[16] 杨庆, 陈桂明, 何庆飞. 局部切空间排列算法用于轴承早期故障诊断[J]. 振动、测试与诊断, 2012, 32(5): 831-835.
Yang Qing, Chen Guiming, He Qingfei. Incipient fault diagnosis of rolling bearings based on local tangent space alignment[J]. Journal of Vibration, Measurement & Diagnosis,  2012, 32(5): 831-835.
[17] 杨正永, 王昕, 王振雷. 基于 LTSA 和联合指标的非高斯过程监控方法及应用[J]. 化工学报, 2015, 66(4): 1370-1379.
Yang Zhengyong, Wang Xin, Wang Zhenlei. LTSA and combined index based non-Gaussian process monitoring and application[J]. CIESC Journal, 2015, 66(4): 1370-1379.
[18] 宋涛, 汤宝平, 李锋. 基于流行学习和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 Baoping, 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.
[19] 孙斌, 刘立远, 牛翀. 基于局部切空间排列和 K-最近邻分类器的转子故障诊断方法[J]. 中国机械工程, 2015, 26(1): 74-78.
Sun Bin, Liu Liyuan, Niu Chong. Rotor fault diagnosis methods based on local tangent space alignment and K-nearest Neighbor[J]. China Machanical Engineering, 2015, 26(1): 74-78.
[20] 许国根, 贾瑛. 模式识别与智能计算的MATLAB实现[M]. 北京: 北京航空航天大学出版社, 2012.
Xu Guogen, Jia Ying. MATLAB implementation of pattern recognition and intelligent computation[M]. Beijing: Beihang University Press, 2012.
[21] 张莉, 孙钢, 郭军. 基于K-均值聚类的无监督的特征选择方法[J]. 计算机应用研究, 2005, 22(3): 23-25.
Zhang Li, Sun Gang, Guo Jun. Unsupervised feature selection method based on K-means clustering[J]. Computer and Modernization, 2005, 22 (3): 23-25.
[22] 杨国安. 信号处理基础[M]. 北京: 中国石化出版社, 2012.
Yang Guoan. Fundamentals of signal processing[M]. Beijing:  China Petrochemical Press, 2012.
[23] 范雪莉, 冯海泓, 原猛. 基于互信息的主成分分析特征选择算法[J]. 控制与决策, 2013, 28(6): 915-919.
FAN Xue-li, FENG Hai-hong, YUAN Meng. PCA based on mutual information for feature selection[J]. Control and Decision, 2013, 28(6): 915-919.
[24] 李月娇, 刘秉瀚. 基于自适应邻域参数的拉普拉斯特征映射[J]. 福州大学学报: 自然科学版, 2013 (2): 153-157.
LI Yuejiao, Liu Binghan. Self-regulation of neighborhood parameter for Laplacian eigenmaps[J]. Journal of Fuzhou University( Natural Science Edition), 2013 (2): 153-157.
[25] 侯臣平, 吴翊, 易东云. 新的流形学习方法统一框架及改进的拉普拉斯特征映射方法[J]. 计算机研究与发展, 2009, 46(4): 676-682.
Hou Chenping, Wu Yi, Yi Dongyun. A novel unified manifold learning framework and an improved Laplacian eigenmaps[J]. Journal of Computer Research and Development, 2009, 46(4): 676-682.

PDF(901 KB)

Accesses

Citation

Detail

段落导航
相关文章

/