Extraction of fault component from abnormal sound in diesel engines using acoustic signals

Abstract In this paper a method for extracting fault components from abnormal acoustic signals and automatically diagnosing diesel engine faults is presented. The method named dislocation superimposed method (DSM) is based on the improved random decrement technique (IRDT), differential function (DF) and correlation analysis (CA). The aim of DSM is to linearly superpose multiple segments of abnormal acoustic signals because of the waveform similarity of faulty components. The method uses sample points at the beginning of time when abnormal sound appears as the starting position for each segment. In this study, the abnormal sound belonged to shocking faulty type; thus, the starting position searching method based on gradient variance was adopted. The coefficient of similar degree between two same sized signals is presented. By comparing with a similar degree, the extracted fault component could be judged automatically. The results show that this method is capable of accurately extracting the fault component from abnormal acoustic signals induced by faulty shocking type and the extracted component can be used to identify the fault type.

[1]  Prem Kumar Kalra,et al.  Audio Signature-Based Condition Monitoring of Internal Combustion Engine Using FFT and Correlation Approach , 2011, IEEE Transactions on Instrumentation and Measurement.

[2]  Jun Li,et al.  Blind Source Separation for Impact Force Induced by Piston Slaps in Diesel Engines , 2013 .

[3]  Ning Wang,et al.  The Research on the Recognition Technique of Rock AE and Noise Signal , 2010 .

[4]  S. R. Ibrahim,et al.  Vector Triggering Random Decrement for High Identification Accuracy , 1998 .

[5]  S. R. Ibrahim,et al.  STATISTICAL THEORY OF THE VECTOR RANDOM DECREMENT TECHNIQUE , 1999 .

[6]  P. Tse,et al.  A comparison study of improved Hilbert–Huang transform and wavelet transform: Application to fault diagnosis for rolling bearing , 2005 .

[7]  Babak Khodabandeloo,et al.  A random decrement technique for operational modal analysis in the presence of periodic excitations , 2009 .

[8]  Yi Min Shao,et al.  Engine Assembly Quality Defect Inspection Based on EMD-Envelope Spectra , 2013 .

[9]  Xia Wang,et al.  Fault diagnosis of diesel engine based on adaptive wavelet packets and EEMD-fractal dimension , 2013 .

[10]  M.J. Mahjoob,et al.  Application of Discrete Wavelet Transform (DWT) in Combustion Failure Detection of IC Engines , 2007, 2007 5th International Symposium on Image and Signal Processing and Analysis.

[11]  M. Muñoz,et al.  Engine diagnosis method based on vibration and acoustic emission energy , 2012 .

[12]  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.

[13]  S. M. Jafari,et al.  Valve Fault Diagnosis in Internal Combustion Engines Using Acoustic Emission and Artificial Neural Network , 2014 .

[14]  Katsumi Yamashita,et al.  A theoretical discussion on the foundation of Stone’s blind source separation , 2011, Signal Image Video Process..

[15]  Inés Couso,et al.  Interval-valued Blind Source Separation Applied to AI-based Prognostic Fault Detection of Aircraft Engines , 2014, J. Multiple Valued Log. Soft Comput..

[16]  Antonio Paolo Carlucci,et al.  Analysis of the relation between injection parameter variation and block vibration of an internal combustion diesel engine , 2006 .

[17]  Bogusław Łazarz,et al.  Condition monitoring of engine timing system by using wavelet packet decomposition of a acoustic signal , 2014 .

[18]  Teng We.  Semi-speed oil whirl fault component extraction in a steam turbine based on time-frequency filtering , 2015 .

[19]  Piotr Czech,et al.  Application of the discrete wavelet transform and probabilistic neural networks in IC engine fault diagnostics , 2014 .

[20]  Yu Yang,et al.  Partly ensemble empirical mode decomposition: An improved noise-assisted method for eliminating mode mixing , 2014, Signal Process..