Fault diagnosis of diesel engine based on adaptive wavelet packets and EEMD-fractal dimension

Abstract In this paper a novel method for de-noising nonstationary vibration signal and diagnosing diesel engine faults is presented. The method is based on the adaptive wavelet threshold (AWT) de-noising, ensemble empirical mode decomposition (EEMD) and correlation dimension (CD). A new adaptive wavelet packet (WP) thresholding function for vibration signal de-noising is used in this paper. To alleviate the mode mixing problem occurring in EMD, ensemble empirical mode decomposition (EEMD) is presented. With EEMD, the components with truly physical meaning can be extracted from the signal. Utilizing the advantage of EEMD, this paper proposes a new AWT–EEMD-based method for fault diagnosis of diesel engine. A study of correlation dimension in engine condition monitoring is reported also. Some important influencing factors relating directly to the computational precision of correlation dimension are discussed. Industrial engine normal and fault vibration signals measured from different operating conditions are analyzed using the above method. These techniques have integrated with our proposed adaptive wavelet threshold de-noising to form a new AWT–EEMD–CD method. The advantage of combining of EEMD and fractal dimension is that it does not require the classifiers to recognize the diesel engine fault types, also the method can solve the difficulty of recognizing fault states when two or more fractal dimensions are close to each other. To verify the effectiveness of the EEMD–CD in detecting the faults, their induced vibrations are collected from a series of generators under normal and faulty engine conditions after de-noising. The results show that this method is capable of extracting the impact signal features induced by vibrations and is able to determine their types of fault accurately even when the impacts have been overwhelmed by other unrelated vibration signals.

[1]  Gabriel Rilling,et al.  EMD Equivalent Filter Banks, from Interpretation to Applications , 2005 .

[2]  Norden E. Huang,et al.  Ensemble Empirical Mode Decomposition: a Noise-Assisted Data Analysis Method , 2009, Adv. Data Sci. Adapt. Anal..

[3]  Zhiquan Qi,et al.  A New Support Vector Machine for Multi-class Classification , 2005, CIT.

[4]  Jien-Chen Chen,et al.  Continuous wavelet transform technique for fault signal diagnosis of internal combustion engines , 2006 .

[5]  Yi-Hsing Tseng,et al.  Feature extraction of hyperspectral data using the Best Wavelet Packet Basis , 2002, IEEE International Geoscience and Remote Sensing Symposium.

[6]  Ronald R. Coifman,et al.  Local discriminant bases and their applications , 1995, Journal of Mathematical Imaging and Vision.

[7]  Jian Chen,et al.  THE APPLICATION OF CORRELATION DIMENSION IN GEARBOX CONDITION MONITORING , 1999 .

[8]  Richard David Neilson,et al.  THE USE OF CORRELATION DIMENSION IN CONDITION MONITORING OF SYSTEMS WITH CLEARANCE , 2000 .

[9]  Ethem Alpaydin,et al.  Support Vector Machines for Multi-class Classification , 1999, IWANN.

[10]  L. Cao.  Practical method for determining the minimum embedding dimension of a scalar time series , 1997 .

[11]  Jan Larsen,et al.  Condition monitoring with Mean field independent components analysis , 2005 .

[12]  Klaus Fraedrich,et al.  Estimating the correlation dimension of an attractor from noisy and small datasets based on re-embedding , 1993 .

[13]  Joseph Mathew,et al.  USING THE CORRELATION DIMENSION FOR VIBRATION FAULT DIAGNOSIS OF ROLLING ELEMENT BEARINGS—I. BASIC CONCEPTS , 1996 .

[14]  Ting Wu,et al.  Adaptive subject-based feature extraction in brain-computer interfaces using wavelet packet best basis decomposition. , 2007, Medical engineering & physics.

[15]  Junyan Yang,et al.  Intelligent fault diagnosis of rolling element bearing based on SVMs and fractal dimension , 2007 .

[16]  David L. Donoho,et al.  De-noising by soft-thresholding , 1995, IEEE Trans. Inf. Theory.

[17]  D. Kugiumtzis.  State space reconstruction parameters in the analysis of chaotic time series—the role of the time window length , 1996, comp-gas/9602002.

[18]  N. Huang,et al.  The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis , 1998, Proceedings of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences.

[19]  P. Grassberger,et al.  Characterization of Strange Attractors , 1983 .

[20]  Chi-Man Vong,et al.  Engine ignition signal diagnosis with Wavelet Packet Transform and Multi-class Least Squares Support Vector Machines , 2011, Expert Syst. Appl..

[21]  Zhongsheng Wang,et al.  Robust incipient fault identification of aircraft engine rotor based on wavelet and fraction , 2010 .

[22]  Pizhong Qiao,et al.  On the wavelet–fractal nonlinear damage diagnosis of mechanical systems , 2009 .

[23]  Yuichi Matsumura,et al.  Diagnosis with the Correlation Integral in Time Domain , 2000 .

[24]  Joseph Mathew,et al.  USING THE CORRELATION DIMENSION FOR VIBRATION FAULT DIAGNOSIS OF ROLLING ELEMENT BEARINGS—II. SELECTION OF EXPERIMENTAL PARAMETERS , 1996 .

[25]  P. Grassberger,et al.  Measuring the Strangeness of Strange Attractors , 1983 .

[26]  J. Salas,et al.  Nonlinear dynamics, delay times, and embedding windows , 1999 .

[27]  S. S. Shen,et al.  A confidence limit for the empirical mode decomposition and Hilbert spectral analysis , 2003, Proceedings of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences.

[28]  I. Johnstone,et al.  Ideal spatial adaptation by wavelet shrinkage , 1994 .