在强背景噪声和复杂激励的干扰下，滚动轴承的早期微弱故障特征往往难以提取，提出一种稀疏分解与频域相关峭度相结合的方式，对轴承早期微弱故障特征进行提取。稀疏表示方法是分析非平稳信号的一种有效方式，在轴承故障诊断中常用的一种方法是利用K-SVD算法构造自适应字典，采用OMP算法对采集到的数据进行稀疏分解。利用频域相关峭度能够准确识别出轴承等旋转机械的循环冲击序列的特性，将其引入到字典构造过程中，求解稀疏分解时每次迭代逼近信号的频域相关峭度，并且找到最大频域相关峭度值所在位置，根据当前位置的信号重构原始信号，计算其包络及包络谱，分析故障类型。仿真信号和试验信号的结果表明：所提方法能够准确识别出轴承故障，验证了该方法在识别循环冲击序列的有效性和优越性。
Abstract
Under interference of strong background noise and complex excitation, early weak fault features of rolling bearing are often difficult to extract.The sparse representation method is an effective way to analyze non-stationary signals, and adopting K-SVD algorithm to construct an adaptive dictionary and OMP algorithm to sparsely decompose the acquired data is a common method in bearing fault diagnosis.Here, a method combining sparse decomposition and frequency domain correlation kurtosis was proposed to extract bearings’ early weak fault features.The frequency domain correlation kurtosis with the advantage to be able to accurately recognize bearing, etc.rotating machineries’ cyclic impact sequence features was used to construct the adaptive dictionary.Firstly, the frequency domain correlation kurtosis of the signal approached by each iteration was solved when performing sparse decomposition.Secondly, the position for the maximum frequency domain correlation kurtosis value was found.Finally, the signal corresponding to this position was used to reconstruct the original signal, calculate its envelope and envelope spectrum, and analyze bearing fault types.The analysis results of simulated signals and ones obtained in tests showed that the proposed method can be used to accurately identify bearing faults, and verify the effectiveness and superiority of this method in identifying cyclic impact sequences.
关键词
滚动轴承 /
微弱故障 /
稀疏分解 /
循环平稳
{{custom_keyword}} /
Key words
rolling bearing /
weak fault /
sparse decomposition /
cyclic stationary
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] MALLAT S G, ZHANG Z. Matching pursuits with time-frequency dictionaries[J]. IEEE Transactions on signal processing, 1993, 41(12): 3397-3415.
[2] 王林,蔡改改,高冠琪,等.基于改进MP的稀疏表示快速算法及其滚动轴承故障特征提取应用[J].振动与冲击,2017,36(03):176-182.
WANG Lin, CAI Gaigai, GAO Guanqi, et al. Fast sparse representation algorithm based on improved MP and its applications fault feature extraction of rolling bearings [J]. Journal of Vibration and Shock, 2017,36 (03): 176-182.
[3] 樊薇,李双,蔡改改,等.基于小波基稀疏信号特征提取的轴承故障诊断[J].振动工程学报,2015,28(06):972-980.
FAN Wei, LI Shuang, CAI Gaigai, et al. Wavelet sparse signal feature extracion method and its application in bearing fault diagnosis [J]. Journal of vibration engineering, 2015,28 (06): 972-980.
[4] 樊薇,李双,蔡改改,沈长青,黄伟国,朱忠奎.瞬态成分Laplace小波稀疏表示及其轴承故障特征提取应用[J].机械工程学报,2015,51(15):110-118.
FAN Wei, LI Shuang, CAI Gaigai, et al. Sparse representation for transient in Laplace wavelet basis and its application in feature extraction of bearing fault [J]. Journal of mechanical engineering, 2015,51 (15): 110-118.
[5] HE W, DING Y, ZI Y, et al. Sparsity-based algorithm for detecting faults in rotating machines[J]. Mechanical Systems and Signal Processing, 2016, 72: 46-64.
[6] ZHANG H, CHEN X, DU Z, et al. Kurtosis based weighted sparse model with convex optimization technique for bearing fault diagnosis[J]. Mechanical Systems and Signal Processing, 2016, 80: 349-376.
[7]郭俊锋,石斌,魏兴春,等.基于K-SVD字典学习算法的稀疏表示振动信号压缩测量重构方法[J].机械工程学报,2018,54(07):97-106.
GUO Junfeng, SHI Bin, WEI Xingchun, et al. A method of reconstruction of compressed measuring for mechinical vibration signals based on K-SVD dictionary-training algorithm sparse representation [J]. Journal of mechanical engineering, 2018,54 (07): 97-106.
[8] WANG D, ZHAO Y, YI C, et al. Sparsity guided empirical wavelet transform for fault diagnosis of rolling element bearings[J]. Mechanical Systems and Signal Processing, 2018, 101: 292-308.
[9] DING J, ZHAO W, MIAO B, et al. Adaptive sparse representation based on circular-structure dictionary learning and its application in wheelset-bearing fault detection[J]. Mechanical Systems and Signal Processing, 2018, 111: 399-422.
[10] MCDONALD G L，ZHAO Q，ZUO M J．Maximum correlated kurtosis deconvolution and application on gear tooth chip fault detection[J]．Mechanical Systems and Signal Processing，2012，33：237-255．
[11] ANTONI J．The infogram：Entropic evidence of the signature of repetitive transients[J]．Mechanical Systems and Signal Processing，2016，74：73-94．
[12] 顾晓辉,杨绍普,刘永强,等.一种改进的峭度图方法及其在复杂干扰下轴承故障诊断中的应用[J].振动与冲击,2017,36(23):187-193.
GU Xiaohui, YANG Shaopu, LIU Yongqiang, et al. An improved kurtogram method and its application in fault diagnosis of rolling element bearings under complex interferences [J].Journal of Vibration and Shock,2017,36(23): 187-193.
[13] DONOHO, DAVID L. Compressed sensing. IEEE Transactions on information theory, 2006, 52.4: 1289-1306.
[14] CANDÈS, Emmanuel J., ROMBERG, et al. Robust uncertainty principles: Exact signal reconstruction from highly incomplete frequency information. IEEE Transactions on information theory, 2006, 52.2: 489-509.
[15] CANDÈS, EMMANUEL J., et al. Compressive sampling. In: Proceedings of the international congress of mathematicians. 2006. p. 1433-1452.
[16] ENGAN, KJERSTI; AASE, et al. Method of optimal directions for frame design. In: Acoustics, Speech, and Signal Processing, 1999. Proceedings., 1999 IEEE International Conference on. IEEE, 1999. p. 2443-2446.
[17] MAIRAL J., BACH F., PONCE J, et al. Online learning for matrix factorization and sparse coding. Journal of Machine Learning Research, 2010, 11.Jan: 19-60.
[18] AHARON M, ELAD M, BRUCKSTEIN A. rm k-SVD: An algorithm for designing overcomplete dictionaries for sparse representation[J]. IEEE Transactions on signal processing, 2006, 54(11): 4311-4322.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
PDF(1370 KB)
