Abstract：A new method for diagnosis based on ensemble empirical mode of decomposition(EEMD), singularity value decomposition(SVD) and fuzzy C-means clustering(FCM) is proposed. First of all, the mechanical vibration signals were decomposed by EEMD into a certain number of intrinsic mode functions (IMFs). Secondly, IMF components were chosen by the criteria of mutual correlation coefficient and got several component containing the main information of signals, then , with the SVD method ,singular value sequences were obtained. At last, the constructed eigenvector were put into the FCM fuzzy clustering classifier to recognize different fault types. The results of experiment and engineering analysis demonstrate that the method proposed is able to diagnose mechanical faults accurately and effectively.
［1］Huang N E, Shen Z, Long S R. The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis［J］. Proc R Soc, 1998, 454: 903-995.
［2］Huang N E, Long S R. A new view of nonlinear water wave: the Hilbert spectrum ［J］. Ann Rev Fluid Mech, 1999, 31: 417-457.
［3］Wu Zhaohua, Huang N E. Ensemble empirical mode decomposition: A noise-assisted data analysis method［J］. Advances in Adaption Data Analysis, 2009, 1(1): 1-41.
［4］胡广书.数字信号处理—理论、算法与实现［M］.北京:清华大学出版社,2003.
［5］张贤达. 矩阵分析与应用［M］.哈尔滨:哈尔滨工业大学出版社,2004, 169-172.
［6］高琦, 崔长彩, 胡捷,等. 基于模糊C均值聚类算法的金刚石砂轮磨粒边缘检测［J］. 计量学报, 2014,35(4): 315-322.
［7］戴桂平. 基于EMD近似熵和DAGSVM的机械故障诊断［J］. 计量学报, 2010, 31(5): 467-471.
2016, Vol. 37 
