VMD-LSSVM Fault Identification Method for Rolling Bearings
-
摘要: 滚动轴承是工程设备中的关键部件,对滚动轴承进行故障识别方法研究有重要的意义。为了解决滚动轴承振动信号分析能力薄弱的问题,提出了一种基于变分模态分解(Variational modedecomposition,VMD)与最小二乘支持向量机(Least square support vector machine,LSSVM)的滚动轴承故障识别方法。以凯斯西储大学滚动轴承实验数据为研究对象,获取4类故障7种滚动轴承状态实验振动数据。进行VMD分解,得出最佳分解本征模态函数(Intrinsic mode function,IMF)个数4,然后计算4个IMF样本熵(Sample entropy,SE)得到相应特征量,输入LSSVM模型进行状态识别。实验表明,基于VMD-LSSVM的方法比EMD(Empirical mode decomposition)-HMM(HiddenMarkov model)和EMD-LSSVM方法有更高的识别率。Abstract: Rolling bearing is a key component in engineering equipment, the research on fault identification method of rolling bearings is of great significance. In order to solve the problem of weak ability of rolling bearing vibration signal analysis, a rolling bearing fault identification method based on variational mode decomposition (VMD) and least square support vector machine (LSSVM) is proposed. Taking the rolling bearing experimental data of Case Western Reserve University as the research object, the experimental vibration data of four types of faults and seven rolling bearing states are obtained. VMD decomposition is carried out to obtain the number 4 of the best decomposed intrinsic mode function (IMF), and then the four IMF sample entropy (SE) is calculated to obtain the corresponding feature quantity, which is input into the LSSVM model for state recognition. Experiments show that the proposed method based on VMD-LSSVM has higher recognition rate than EMD (empirical mode decomposition) - HMM (Hidden Markov model) and EMD-LSSVM.
-
Key words:
- rolling bearing /
- fault diagnosis /
- VMD /
- SE /
- LSSVM /
- IMF
-
表 1 滚动轴承数据集
Table 1. Rolling bearing dataset
标签 故障类型 故障直径/mm 1 正常 0 2 内圈 0.1778 3 滚动体 0.1778 4 外圈 0.1778 5 内圈 0.3556 6 滚动体 0.3556 7 外圈 0.3556 表 2 取不同模态数时VMD分解中心频率
Table 2. Center frequencies in VMD decomposition withdifferent modes
模态数k IMF1 IMF2 IMF3 IMF4 IMF5 2 1270.8 3276.0 − − − 3 576.0 2674.8 3574.8 − − 4 572.4 1359.6 2707.2 3576 − 5 570.0 1347.6 2686.8 3354 3642 表 3 部分故障特征数据集
Table 3. Partial fault feature dataset
状态标签 样本序列 特征向量 T1 T2 T3 T4 1 1 0.0267 0.3252 0.4016 0.5934 2 0.0187 0.3257 0.4355 0.9453 3 0.1142 0.3383 0.4183 0.6038 2 4 0.4164 0.4242 0.5489 0.2776 5 0.4268 0.4167 0.5291 0.2949 6 0.4559 0.4180 0.5594 0.2878 3 7 0.3787 0.5793 0.6715 0.1965 8 0.3393 0.6320 0.6766 0.1992 9 0.3510 0.5458 0.6936 0.1784 4 10 0.7256 0.0330 0.1172 0.1384 11 0.7124 0.0304 0.1049 0.1445 12 0.6507 0.0283 0.1174 0.1088 5 13 0.4078 0.1912 0.3854 0.2052 14 0.4456 0.2125 0.4294 0.2103 15 0.4480 0.1661 0.3895 0.2048 6 16 0.3611 0.3262 0.7465 0.2535 17 0.3610 0.3186 0.7142 0.2301 18 0.3603 0.3550 0.5346 0.1247 7 19 0.4319 0.5137 0.9939 0.5508 20 0.4871 0.8415 0.7525 0.5740 21 0.4510 0.4964 0.9669 0.5080 表 4 3种方法运行20次平均识别率
Table 4. Average recognition rates of the three methodsover 20 runs
方法 平均识别率/% EMD-HMM 76.357 EMD-LSSVM 79.571 VMD-LSSVM 95.607 -
[1] ZHAO H M, SUN M, DENG W, et al. A new feature extraction method based on EEMD and multi-scale fuzzy entropy for motor bearing[J]. Entropy, 2017, 19(1): 14. [2] WANG X L, HAN W H, GU H, et al. Research on rolling element bearing fault diagnosis based on EEMD and correlated kurtosis[J]. Applied Mechanics and Materials, 2014, 680: 198-205. doi: 10.4028/www.scientific.net/AMM.680.198 [3] 甄冬, 朱继瑞, 张琛, 等. 基于小波包能量与峭度谱的滚动轴承故障诊断[J]. 机械设计, 2021, 38(2): 23-28.ZHEN D, ZHU J R, ZHANG C, et al. Fault diagnosis of rolling bearing based on wavelet packet energy and spectral kurtosis[J]. Journal of Machine Design, 2021, 38(2): 23-28. (in Chinese) [4] 吴宏亮, 尚坤. 基于EMD分解及Hilbert包络的电机轴承故障诊断[J]. 电子技术, 2021, 50(7): 112-115.WU H L, SHANG K. Motor fault diagnosis based on EMD decomposition and Hilbert envelope[J]. Electronic Technology, 2021, 50(7): 112-115. (in Chinese) [5] 朱杰平, 张永祥, 王孝霖, 等. 基于EMD分解和相关峭度的滚动轴承故障诊断方法研究[J]. 武汉理工大学学报(交通科学与工程版), 2014, 38(2): 367-370.ZHU J P, ZHANG Y X, WANG X L, et al. Research on rolling element bearing fault diagnosis based on empirical mode decomposition and correlated kurtosis[J]. Journal of Wuhan University of Technology (Transportation Science & Engineering), 2014, 38(2): 367-370. (in Chinese) [6] WU Z H, HUANG N E. Ensemble empirical mode decomposition: a noise-assisted data analysis method[J]. Advances in Adaptive Data Analysis, 2009, 1(1): 1-41. doi: 10.1142/S1793536909000047 [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, 28(2): 324-330. doi: 10.16385/j.cnki.issn.1004-4523.2015.02.020ZHENG Z, JIANG W L, HU H S, et al. Research on rolling bearings fault diagnosis method based on EEMD morphological spectrum and kernel fuzzy C-means clustering[J]. Journal of Vibration Engineering, 2015, 28(2): 324-330. (in Chinese) doi: 10.16385/j.cnki.issn.1004-4523.2015.02.020 [8] COLOMINAS M A, SCHLOTTHAUER G, TORRES M E. Improved complete ensemble EMD: a suitable tool for biomedical signal processing[J]. Biomedical Signal Processing and Control, 2014, 14: 19-29. doi: 10.1016/j.bspc.2014.06.009 [9] 丁承君, 张良, 冯玉伯, 等. VMD和t-SNE相结合的滚动轴承故障诊断[J]. 机械科学与技术, 2020, 39(5): 758-764. doi: 10.13433/j.cnki.1003-8728.20190193DING C J, ZHANG L, FENG Y B, et al. Fault diagnosis method of rolling bearing combining VMD with t-SNE[J]. Mechanical Science and Technology for Aerospace Engineering, 2020, 39(5): 758-764. (in Chinese) doi: 10.13433/j.cnki.1003-8728.20190193 [10] 董治麟, 郑近德, 潘海洋, 等. 一种时移多尺度排列熵与ELM相结合的滚动轴承故障诊断方法[J]. 机械科学与技术, 2021, 40(10): 1523-1529.DONG Z L, ZHENG J D, PAN H Y, et al. A rolling bearing fault diagnosis method of time-shifted multi-scale permutation entropy combining with ELM[J]. Mechanical Science and Technology for Aerospace Engineering, 2021, 40(10): 1523-1529. (in Chinese) [11] 张朝林, 范玉刚, 冯早. ITD-多尺度熵和ELM的风电轴承健康状态识别[J]. 机械科学与技术, 2018, 37(11): 1731-1736. doi: 10.13433/j.cnki.1003-8728.20180121ZHANG Z L, FAN Y G, FENG Z. Health status recognition of wind turbine bearings based on ITD-MSE and ELM[J]. Mechanical Science and Technology for Aerospace Engineering, 2018, 37(11): 1731-1736. (in Chinese) doi: 10.13433/j.cnki.1003-8728.20180121 [12] 邹龙庆, 陈桂娟, 邢俊杰, 等. 基于LMD样本熵与SVM的往复压缩机故障诊断方法[J]. 噪声与振动控制, 2014, 34(6): 174-177. doi: 10.3969/j.issn.1006-1355.2014.06.039ZOU L Q, CHEN G J, XING J J, et al. Fault diagnosis method based on LMD sample entropy and SVM for reciprocating compressors[J]. Noise and Vibration Control, 2014, 34(6): 174-177. (in Chinese) doi: 10.3969/j.issn.1006-1355.2014.06.039 [13] 赵元喜, 胥永刚, 高立新, 等. 基于谐波小波包和BP神经网络的滚动轴承声发射故障模式识别技术[J]. 振动与冲击, 2010, 29(10): 162-165. doi: 10.3969/j.issn.1000-3835.2010.10.033ZHAO Y X, XU Y G, GAO L X, et al. Fault pattern recognition technique for roller bearing acoustic emission based on harmonic wavelet packet and BP neural network[J]. Journal of Vibration and Shock, 2010, 29(10): 162-165. (in Chinese) doi: 10.3969/j.issn.1000-3835.2010.10.033 [14] 李庆, LIANG S Y, 杨建国. 谐波小波样本熵与HMM模型的轴承故障模式识别[J]. 上海交通大学学报, 2016, 50(5): 723-729.LI Q, LIANG S Y, YANG J G. Bearing fault pattern recognition using harmonic wavelet sample entropy and hidden Markov model[J]. Journal of Shanghai Jiaotong University, 2016, 50(5): 723-729. (in Chinese) [15] 韩美东, 张金豹, 赵永强. ALIF-MMPE结合DAG-SVM的滚动轴承故障诊断[J]. 机械科学与技术, 2020, 39(9): 1358-1365. doi: 10.13433/j.cnki.1003-8728.20190272HAN M D, ZHANG J B, ZHAO Y Q. Rolling bearing fault diagnosis with ALIF-MMPE and DAG-SVM[J]. Mechanical Science and Technology for Aerospace Engineering, 2020, 39(9): 1358-1365. (in Chinese) doi: 10.13433/j.cnki.1003-8728.20190272 [16] 熊景鸣, 潘林, 朱昇, 等. DBN与PSO-SVM的滚动轴承故障诊断[J]. 机械科学与技术, 2019, 38(11): 1726-1731. doi: 10.13433/j.cnki.1003-8728.20190040XIONG J M, PAN L, ZHU S, et al. Bearing fault diagnosis based on deep belief networks and particle swarm optimization support vector machine[J]. Mechanical Science and Technology for Aerospace Engineering, 2019, 38(11): 1726-1731. (in Chinese) doi: 10.13433/j.cnki.1003-8728.20190040 [17] 王普, 辛娇娇, 高学金, 等. 基于独立元分析-最小二乘支持向量机的冷水机组故障诊断方法[J]. 北京工业大学学报, 2017, 43(11): 1641-1647. doi: 10.11936/bjutxb2016070012WANG P, XIN J J, GAO X J, et al. Fault diagnosis of chiller based on independent component analysis and least squares support vector machine[J]. Journal of Beijing University of Technology, 2017, 43(11): 1641-1647. (in Chinese) doi: 10.11936/bjutxb2016070012 [18] 唐勇波, 熊印国. 相对变换主元分析特征提取的变压器故障诊断[J]. 系统仿真学报, 2018, 30(3): 1127-1133. doi: 10.16182/j.issn1004731x.joss.201803045TANG Y B, XIONG Y G. Transformer fault diagnosis based on feature extraction of relative transformation principal component analysis[J]. Journal of System Simulation, 2018, 30(3): 1127-1133. (in Chinese) doi: 10.16182/j.issn1004731x.joss.201803045 [19] CHEN F, CHEN X J, YANG Z J, et al. A rolling bearing fault diagnosis method based on VMD-multiscale fractal dimension/energy and optimized support vector machine[J]. Journal of Vibroengineering, 2016, 18(6): 3581-3595. doi: 10.21595/jve.2016.16847 [20] PANDEY A, SAINI B S, SINGH B, et al. A 2D electrocardiogram data compression method using a sample entropy-based complexity sorting approach[J]. Computers & Electrical Engineering, 2016, 56: 30-45. [21] HAN M H, PAN J L. A fault diagnosis method combined with LMD, sample entropy and energy ratio for roller bearings[J]. Measurement, 2015, 76: 7-19. doi: 10.1016/j.measurement.2015.08.019 [22] 常春, 梅检民, 赵慧敏, 等. 基于局部切空间排列和最小二乘支持向量机的气缸压力识别[J]. 振动与冲击, 2020, 39(13): 16-21. doi: 10.13465/j.cnki.jvs.2020.13.003CHANG C, MEI J M, ZHAO H M, et al. Recognition of cylinder pressure based on LTSA-LSSVM[J]. Journal of Vibration and Shock, 2020, 39(13): 16-21. (in Chinese) doi: 10.13465/j.cnki.jvs.2020.13.003 [23] SEDAGHAT M, ROUHIBAKHSH K. On prediction of asphaltene precipitation in different operational conditions utilization of LSSVM algorithm[J]. Petroleum Science and Technology, 2018, 36(16): 1292-1297. doi: 10.1080/10916466.2018.1471491 [24] 马增强, 李亚超, 刘政, 等. 基于变分模态分解和Teager能量算子的滚动轴承故障特征提取[J]. 振动与冲击, 2016, 35(13): 134-139. doi: 10.13465/j.cnki.jvs.2016.13.022MA Z Q, LI Y C, LIU Z, et al. Rolling bearings' fault feature extraction based on variational mode decomposition and Teager energy operator[J]. Journal of Vibration and Shock, 2016, 35(13): 134-139. (in Chinese) doi: 10.13465/j.cnki.jvs.2016.13.022 [25] 郑近德, 刘涛, 孟瑞, 等. 基于广义复合多尺度排列熵与PCA的滚动轴承故障诊断方法[J]. 振动与冲击, 2018, 37(20): 61-66. doi: 10.13465/j.cnki.jvs.2018.20.010ZHENG J D, LIU T, MENG R, et al. Generalized composite multiscale permutation entropy and PCA based fault diagnosis of rolling bearings[J]. Journal of Vibration and Shock, 2018, 37(20): 61-66. (in Chinese) doi: 10.13465/j.cnki.jvs.2018.20.010 [26] 冷军发, 牛振华, 荆双喜, 等. EMD与CICA方法在多级齿轮传动微弱故障特征提取中的应用[J]. 机械科学与技术, 2017, 36(7): 1029-1034. doi: 10.13433/j.cnki.1003-8728.2017.0708LENG J F, NIU Z H, JING S X, et al. Weak fault feature extraction of multi-stage gear transmission based on EMD and CICA[J]. Mechanical Science and Technology, 2017, 36(7): 1029-1034. (in Chinese) doi: 10.13433/j.cnki.1003-8728.2017.0708