Rolling Bearing Fault Diagnosis with ALIF-MMPE and DAG-SVM
-
摘要: 针对滚动轴承故障诊断中非平稳振动信号下的有效故障特征提取问题,提出一种基于自适应局部迭代滤波、多元多尺度排列熵和有向无环图算法支持向量机的滚动轴承故障诊断方法。自适应局部迭代滤波通过构建自适应滤波函数,能够有效抑制噪声和模态混叠,经自适应分解后得到若干本征模态函数。仿真结果表明其效果优于经验模态分解。然后利用多元多尺度排列熵对包含显著故障信息的本征模态函数进行信息融合和特征提取,组成故障状态特征集。采用主成分分析对故障状态特征集进行降维,随机抽取部分样本带入有向无环图算法支持向量机中进行训练,其它则作为测试样本进行故障识别和诊断。试验故障诊断结果表明:自适应局部迭代滤波下多元多尺度排列熵优于多个本征模态函数下的多尺度排列熵和经验模态分解下的多元多尺度排列熵;本文方法能准确地识别滚动轴承不同的故障类型及故障程度。
-
关键词:
- 轴承故障诊断 /
- 自适应局部迭代滤波 /
- 多元多尺度排列熵 /
- 有向无环图算法支持向量机
Abstract: An approach based on adaptive local iterative filtering (ALIF), multivariate multiscale permutation entropy (MMPE) and directed acyclic graph support vector machine (DAG-SVM) is proposed in this paper for the problem of efficient fault feature extraction from the non-stationary vibration signals in the fault diagnosis of rolling bearings. ALIF could construct adaptive filtering functions to efficiently constrain noise and mode mixing, and a number of intrinsic mode functions (IMFs) are obtained with adaptive decomposition. The simulation result demonstrates that ALIF outperforms empirical mode decomposition (EMD). Then MMPE is employed to perform information infusion and feature extraction with the IMFs containing primary faulty information, and fault state feature sets are constructed with MMPE. After dimensionality reduction using principal component analysis (PCA), some of the new features are randomly selected to train the DAG-SVM, while the remaining ones are used as test samples for identification and fault diagnosis. Results of the experiment show that MMPE of ALIF is superior to the multiscale permutation entropy (MPE) of multiple IMFs and MMPE of EMD; the proposed method could efficiently identify different types of faults and severity levels of rolling bearings. -
表 1 不同组别下的轴承故障诊断
组别 类型 损伤程度/inch A 滚动体、内圈、外圈 0, 0.007 B 滚动体 0, 0.007, 0.014, 0.021, 0.028 C 内圈 0, 0.007, 0.014, 0.021, 0.028 D 外圈 0, 0.007, 0.014, 0.021 表 2 不同组别下的故障识别准确率(%)
组别 EMD MMPE ALIF (IMF1~3)-MPE ALIF MMPE A 95.25(1.96) 97.44(1.63) 98.87(0.82) B 88.50(2.60) 82.70(2.00) 92.25(1.96) C 85.85(3.34) 97.90(1.22) 98.65(1.84) D 90.50(2.96) 98.31(2.36) 96.12(2.18) -
[1] 党建, 贾嵘, 罗兴锜, 等.基于H-K聚类逻辑回归的贯流式机组水导轴承磨损性能评估研究[J].水利学报, 2017, 48(2):226-233 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=slxb201702012Dang J, Jia R, Luo X Q, et al. Research on wear properties assessment of tubular turbine guide bearing based on H-K clustering-logistic regression model[J]. Shuili Xuebao, 2017, 48(2):226-233(in Chinese) http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=slxb201702012 [2] 施莹, 庄哲, 林建辉.基于EEMD排列熵的高速列车轮对轴承故障诊断方法[J].中国测试, 2017, 43(11):89-95 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=zgcsjs201711018Shi Y, Zhuang Z, Lin J H. Fault diagnosis method of high speed train axle bearing based on EEMD permutation entropy[J]. China Measurement & Test, 2017, 43(11):89-95(in Chinese) http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=zgcsjs201711018 [3] 姚德臣, 杨建伟, 程晓卿, 等.基于多尺度本征模态排列熵和SA-SVM的轴承故障诊断研究[J].机械工程学报, 2018, 54(9):168-176 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jxgcxb201809021Yao D C, Yang J W, Cheng X Q, et al. Railway rolling bearing fault diagnosis based on muti-scale IMF permutation entropy and SA-SVM classifier[J]. Journal of Mechanical Engineering, 2018, 54(9):168-176(in Chinese) http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jxgcxb201809021 [4] 程军圣, 于德介, 杨宇.基于EMD和SVM的滚动轴承故障诊断方法[J].航空动力学报, 2006, 21(3):575-580 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=hkdlxb200603025Cheng J S, Yu D J, Yang Y. Fault diagnosis of roller bearings based on EMD and SVM[J]. Journal of Aerospace Power, 2006, 21(3):575-580(in Chinese) http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=hkdlxb200603025 [5] Feng Z P, Zhang D, Zuo M J. Adaptive mode decomposi-tion methods and their applications in signal analysis for machinery fault diagnosis:a review with examples[J]. IEEE Access, 2017, 5:24301-24331 doi: 10.1109/ACCESS.2017.2766232 [6] 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 of the Royal Society A:Mathematical, Physical and Engineering Sciences, 1998, 454(1971):903-995 doi: 10.1098/rspa.1998.0193 [7] Cicone A, Liu J F, Zhou H M. Adaptive local iterative filtering for signal decomposition and instantaneous frequency analysis[J]. Applied and Computational Harmonic Analysis, 2016, 41(2):384-411 http://www.sciencedirect.com/science/article/pii/S1063520316000129 [8] 杨德友, 王博, 蔡国伟.基于改进自适应局部迭代滤波的谐波检测方法研究[J].仪器仪表学报, 2016, 37(10):2274-2281 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=yqyb201610014Yang D Y, Wang B, Cai G W. Study on the harmonic detection method based on improved adaptive local iterative filtering[J]. Chinese Journal of Scientific Instrument, 2016, 37(10):2274-2281(in Chinese) http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=yqyb201610014 [9] 陈保家, 汪新波, 赵春华, 等.基于自适应局部迭代滤波和能量算子解调的滚动轴承故障特征提取[J].南京理工大学学报, 2018, 42(4):445-452 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=njlgdxxb201804009Chen B J, Wang X B, Zhao C H, et al. Fault feature extraction of rolling bearing based on ALIF and energy operator demodulation[J]. Journal of Nanjing University of Science and Technology, 2018, 42(4):445-452(in Chinese) http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=njlgdxxb201804009 [10] An X L, Zeng H T, Yang W W, et al. Fault diagnosis of a wind turbine rolling bearing using adaptive local iterative filtering and singular value decomposition[J]. Transactions of the Institute of Measurement and Control, 2017, 39(11):1643-1648 doi: 10.1177/0142331216644041 [11] Caesarendra W, Tjahjowidodo T. A review of feature extraction methods in vibration-based condition monitoring and its application for degradation trend estimation of low-speed slew bearing[J]. Machines, 2017, 5(4):21 doi: 10.3390/machines5040021 [12] Bandt C, Pompe B. Permutation entropy:a natural complexity measure for time series[J]. Physical Review Letters, 2002, 88(17):174102 doi: 10.1103/PhysRevLett.88.174102 [13] 程军圣, 马兴伟, 杨宇.基于排列熵和VPMCD的滚动轴承故障诊断方法[J].振动与冲击, 2014, 33(11):119-123 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=zdycj201411021Cheng J S, Ma X W, Yang Y. Rolling bearing fault diagnosis method based on permutation entropy and VPMCD[J]. Journal of Vibration and Shock, 2014, 33(11):119-123(in Chinese) http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=zdycj201411021 [14] 郑近德, 程军圣, 杨宇.基于多尺度熵的滚动轴承故障诊断方法[J].湖南大学学报, 2012, 39(5):38-41 http://d.wanfangdata.com.cn/Periodical/hndxxb201205008Zheng J D, Cheng J S, Yang Y. A rolling bearing fault diagnosis approach based on multiscale entropy[J]. Journal of Hunan University, 2012, 39(5):38-41(in Chinese) http://d.wanfangdata.com.cn/Periodical/hndxxb201205008 [15] Aziz W, Arif M. Multiscale permutation entropy of physiological time series[C]//2005 Pakistan Section Multitopic Conference. Karachi, Pakistan: IEEE, 2005: 1-6 [16] Li Y B, Xu M Q, Wei Y, et al. A new rolling bearing fault diagnosis method based on multiscale permutation entropy and improved support vector machine based binary tree[J]. Measurement, 2016, 77:80-94 doi: 10.1016/j.measurement.2015.08.034 [17] Morabito F C, Labate D, La Foresta F, et al. Multivariate multi-scale permutation entropy for complexity analysis of Alzheimer's disease EEG[J]. Entropy, 2012, 14(7):1186-1202 doi: 10.3390/e14071186 [18] Platt J C, Cristianini N, Shawe-Taylor J. Large margin DAGs for multiclass classification[C]//Advances in Neural Information Processing Systems. Denver: MIT Press, 2000: 547-553 [19] Case Western Reserve University. Bearing data center, seeded fault test data[EB/OL].[2018-09-10]. http://csegroups.case.edu/bearingdatacenter/pages/apparatus-procedures