摘要
针对转子高维故障特征识别精度低的问题,提出基于集成经验模态分解(EEMD)能量矩和邻域保持嵌入(NPE)算法相结合的转子故障分类方法。首先利用EEMD对转子系统的振动故障信号进行分解,得到各阶的本征模态分量(IMF)并计算其能量特征向量矩阵,然后应用NPE算法将高维特征集向低维投影,使降维后类内散度最小化及类间分离度最大化,最后将降维后得到的低维特征集输入K近邻分类器进行模式识别。通过双跨度轴承转子试验台的故障特征数据集验证,结果表明该方法能够有效地解决转子故障特征集的降维问题。
Aiming at the low recognition accuracy of high-dimensional fault features of rotating machinery, it proposes a fault classification method based on integration of ensemble empirical mode decomposition(EEMD) and neighborhood preserving embedding(NPE) algorithm. It uses EEMD to decompose the vibration fault signal of the rotor system, obtains the order of the intrinsic modal function components and builds its energy into the feature vector matrix. Then it applies the NPE algorithm in the project the high-dimensional feature set to the low-dimensional projection, maximizes intra-class divergence minimization and maximization of separation between classes, obtains the low-dimensional feature set after putting dimensionality reduction into the K nearest neighbor classifier for pattern recognition. The fault feature dataset of double-span rotor test-bed is used to verify the results, and the results show that this method can effectively solve the dimensionality reduction problem of the rotor fault feature set.
引文
[1] WU Zhaohua, HUANG Norden E. Ensemble empirical mode decomposition: a noise-assisted data analysis method[J]. Advances in Adaptive Data Analysis, 2009(1):1-41.
[2] HUANG N E, SHEN Z, LONG S R, et al. The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis[J].Proceedings Mathematical Physical & Engineering Sciences, 1998, 454:903-995.
[3] CHEN X,ZHANG J.A novel maximum margin neighborhood preserving embedding for face recognition[J].Future Generation Computer Systems, 2012, 28(1):212-217.
[4] HU K, YUAN J. Statistical monitoring of fed-batch process using dynamic multiway neighborhood preserving embedding[J]. Chemometrics & Intelligent Laboratory Systems, 2008, 90(2):195-203.
[5] 王婷. EMD算法研究及其在信号去噪中的应用[D]. 哈尔滨:哈尔滨工程大学, 2010.
[6] 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.
[7] 郑直, 姜万录, 胡浩松,等. 基于 EEMD 形态谱和 KFCM 聚类集成的滚动轴承故障诊断方法研究[J]. 振动工程学报, 2015(2):324-330.
[8] 陈法法, 李冕, 陈保家,等. EEMD能量熵与优化LS-SVM的滚动轴承故障诊断[J]. 组合机床与自动化加工技术, 2016(12):71-75.
[9] 张敏, 崔海龙, 陈曦晖,等. 基于IMF能量矩和HSMM模型的滚动轴承故障诊断方法[J]. 组合机床与自动化加工技术, 2015(10):101-103.
[10] 孙健, 王成华, 杜庆波. 基于小波包能量谱和NPE的模拟电路故障诊断[J]. 仪器仪表学报, 2013, 34(9):2021-2027.
[11] 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.