Extracting Fault Features of R olling Bearing During Speed Variation Based on cICA
-
摘要: 在ICA基础上发展起来的约束独立分量分析(cICA)方法,可根据一定的先验知识生成参考信号以提取选定的独立分量,解决了原ICA算法的次序不确定性问题。将cICA用于滚动轴承故障诊断,能够根据被监测滚动轴承的特征频率等先验信息建立参考信号并实现对其故障振动特征信号的提取。本文将该方法与针对旋转机械变速过程的阶比跟踪技术和滚动轴承包络分析技术相结合,提出了基于cICA的旋转机械变速工作过程滚动轴承早期故障分析方法。该方法首先通过包络提取技术在共振带获得包含故障信息的包络信号,再通过阶比分析中的等角度采样将包络信号转换到角域,在角域建立参考信号,并用cICA实现旋转机械变速过程下滚动轴承故障对应冲击性信号成分的有效提取。仿真和测试试验表明,所提出方法适合于旋转机械升降速等变速过程中的滚动轴承初期故障特征信息提取。Abstract: The constrained independent component analysis (cICA) is derived from ICA, a method that can be ap-plied to extracting interesting independent components by relying on some prior information, thus overcoming the uncertainty of classic ICA. The use of cICA for diagnosing the faults of a rolling bearing can extract the interesting components of the vibration signal of the faulty bearing according to the prior information on the frequencies of fault features of the bearing. The incorporate order tracking and envelope analysis of cICA are used to effectively extract the fault features of incipient rolling element bearing during speed variation in its angle domain. The fault feature extraction method first obtains envelopes in the resonant frequency band, then applies the even-angle increment re-sampling to convert the envelopes from time domain to angle domain and finally constructs the reference signal for cICA in its angle domain. Both the simulation results and test results show that the method has good performance in extracting the incipient fault features of the bearing of a rotational machine during its speed increase and decrease.
-
[1] Lu W，R ajapakse J C．ICA with R eference[J]. Neuro Compu-ting，2006，(69):2244～2257 [2] Lin Z．Morphologically constrained ICA for extracting weak tem-porally correlated signals[J]．Neuro Computing，2008，(71): 1669～1679 [3] Sri K S，R ajapakse J C．Extracting EEG R hythms using ICA-R [A]. Proceedings of the International Joint Conference on Neural Networks[C]，Hong Kong，China: Institute of Electrical and Electronics Engineers Inc，2008:2133～2138 [4] Wang Z Y，Chen J，Dong G M，Zhou Y．Constrained independ-ent component analysis and its application to machine fault diag-nosis[J]. Mechanical Systems and Signal Processing，2011，(25):2501～2512 [5] 王志阳，陈进，肖文斌，周宇．基于约束独立成分分析的滚动 轴承故障诊断[J]．振动与冲击，2012，31(9):118～121 [6] 郭瑜，高艳，郑华文．旋转机械阶比跟踪中的阶比交叠噪声消 除[J]. 振动与冲击，2008，27(10):98～105 [7] 郭瑜，秦树人，汤宝平，纪跃波．基于瞬时频率估计的旋转机 械阶比跟踪[J]. 机械工程学报，2003，39(3):32～36 [8] Guo Y，Kok Kiong Tan．High efficient crossing-order decoupling in Vold-Kalman filtering order tracking based on independent component analysis[J]. Mechanical Systems and Signal Pro-cessing，2010，24 (6): 1756～1766 [9] Guo Y，Kok Kiong Tan．Order-crossing removal in Gabor order tracking by independent component analysis[J]. Journal of Sound and Vibration，2009，325(1-2):471～488 [10] Hyvarinen A．Independent component analysis: algorithms and applications[J]. Neural Computation，2001，13 (7) [11] Antoni J．Blind separation of vibration components: principles and demonstrations[J]. Mechanical Systems and Signal Pro-cessing，2005，19(6):1166～1180 [12] R andalla R B，Antoni J．R olling element bearing diagnosticsa tu-torial[J]. Mechanical Systems and Signal Processing，2011，(25):485～520 [13] Endo H，R andall R B．Enhancement of autoregressive model based gear tooth fault detection technique by the use of minimum entropy deconvolution filter[J]．Mechanical Systems and Sig-nal Processing，2007，(21):906～919 [14] Antoni J．Fast computation of the kurtogram for the detection of transient faults[J]. Mechanical Systems and Signal Process-ing，2007，(21):108～124 [15] 郭瑜，郑华文，高艳，吴涛．基于谱峭度的滚动轴承包络分析[J]. 振动、 测试与诊断，2011，31(4):517～521 -
点击查看大图
计量
- 文章访问数: 163
- HTML全文浏览量: 21
- PDF下载量: 2
- 被引次数: 0
下载:
