Abstract：To extract fault characteristics from the original signal is hard. For this reason, a novel integrated of incipient fault diagnosis method is presented based on the principle of the improved singular value decomposition(SVD) and empirical mode decomposition(EMD). Firstly, based on multi-resolution singular value decomposition, the original signal is decomposed into approximation and detail signals with different resolution. Then the noise in approximation signal is eliminated by using difference spectrum of singular value. The signal after de-noising is decomposed by EMD and a group of Intrinsic Mode Functions (IMF) is obtained. The IMFs were demodulated with Hilbert transform, and envelope spectrum at each band was obtained, through these procedures the faint feature information can be extracted. The effectiveness of this method is confirmed by the experiment of rolling bearing inner race incipient fault.
［1］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］. Proceeding of Royal Society London, 1998, 454(1971): 903-995.
［2］孟宗, 顾海燕, 刘利晖, 等. 基于 EMD 与 AR 谱的轧机主传动系统故障诊断研究［J］. 计量学报, 2011, 32(4): 338-342.
［3］江航, 尚春阳, 高瑞鹏. 基于 EMD 和神经网络的轮轨故障噪声诊断识别方法研究［J］.振动与冲击,2014,33(17):34-38.
［4］Liu Hongxing, Li Jian, Zhao Ying, et al. Improved singular value decomposition technique for detecting and extracting periodic impulse component in a vibration signal ［J］. Chinese Journal of Mechanical Engineering, 2004, 17(3): 340-345.
［5］Czeslaw Cempel. Generalized singular value decomposition in multidimensional condition monitoring of machines-A proposal of comparative diagnostics［J］. Mechanical Systems and Signal Processing, 2009, 23(3): 701-711.
［6］郑堂, 李世平, 罗鹏, 等. 一种基于 SVD 与 EMD 的多频小信号检测方法［J］. 计量技术, 2013 (6): 18-22.
［7］Zhao Xuezhi, Ye Bangyan. Similarity of signal processing effect between Henkel matrix-based SVD and wavelet transform and its mechanism analysis［J］. Mechanical Systems and Signal Processing, 2009, 23(4): 1062-1075.
［8］张超,陈建军,徐亚兰.基于EMD分解和奇异值差分谱理论的轴承故障诊断方法［J］. 振动工程学报,2011, 24(5): 539-545.
［9］王志武.强噪声背景下机械故障微弱信号特征提取方法研究［D］.山西:中北大学, 2014.
［10］赵学智,叶邦彦,陈统坚.多分辨奇异值分解理论及其在信号处理和故障诊断中的应用［J］.机械工程学报,2010, 46(20): 64-75.
2016, Vol. 37 
