Hidden Markov model-based fault diagnostics method in speed-up and speed-down process for rotating machinery

It is very important to ensure that the large rotating machinery operates safely and reliably. The behaviour characteristics of the speed-up and speed-down process in a rotating machinery possess the distinct diagnostic value. The abundant information, non-stationarity, poor repeatability and reproducibility in the speed-up and speed-down process lead to the necessity to find the corresponding approach of feature extraction and fault recognition. The Hidden Markov model (HMM) is very suitable for modelling the dynamic time series, and has a strong capability of pattern classification, especially for a signal with abundant information, non-stationarity, poor repeatability and reproducibility. At the same time, HMM can process the random long sequences in theory. Based on these features, HMM is very suitable for the signal from the speed-up and speed-down process in rotating machinery. As a result, HMM is introduced to the fault diagnosis of rotating machinery, and a new HMM-based approach of the fault diagnosis for the speed-up and speed-down process is proposed. The main idea of the proposed approach is that the feature vectors, which are obtained by the FFT, wavelet transform, bispectrum, etc., are used as fault features, respectively, and the HMMs as the classifiers to recognise the faults of the speed-up and speed-down process in rotating machinery. The experimental results show that the proposed approach is feasible and effective.

[1]  Feng Chang.  Study on Hidden Markov Models for faults diagnosis of rotor machine in the whole run-up process , 2002 .

[2]  Thomas Kailath,et al.  Modern signal processing , 1985 .

[3]  Krishna R. Pattipati,et al.  A hidden Markov model-based algorithm for fault diagnosis with partial and imperfect tests , 2000, IEEE Trans. Syst. Man Cybern. Part C.

[4]  O.A. Abdel Alim,et al.  HMM/NN hybrids for continuous speech recognition , 2001, Proceedings of the Eighteenth National Radio Science Conference. NRSC'2001 (IEEE Cat. No.01EX462).

[5]  Zhao Li,et al.  Study on the Chinese continuous speech recognition under noise environments based on PCANN/HMM , 2003, International Conference on Neural Networks and Signal Processing, 2003. Proceedings of the 2003.

[6]  Kenneth A. Loparo,et al.  A new bearing fault detection and diagnosis scheme based on hidden Markov modeling of vibration signals , 2001, 2001 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings (Cat. No.01CH37221).

[7]  D. Stegemann,et al.  Monitoring and vibrational diagnostic of rotating machinery in power plants , 1998 .

[8]  A. Cohen,et al.  On the use of hidden Markov models in infants' cry classification , 2002, The 22nd Convention on Electrical and Electronics Engineers in Israel, 2002..

[9]  Kee-Choon Kwon,et al.  Accident identification in nuclear power plants using hidden Markov models , 1999 .

[10]  J. Penman,et al.  The use of hidden Markov models for condition monitoring electrical machines , 1993 .

[11]  Padhraic Smyth,et al.  Hidden Markov models and neural networks for fault detection in dynamic systems , 1993, Neural Networks for Signal Processing III - Proceedings of the 1993 IEEE-SP Workshop.

[12]  Lawrence R. Rabiner,et al.  A tutorial on hidden Markov models and selected applications in speech recognition , 1989, Proc. IEEE.

[13]  Jerry M. Mendel,et al.  Tutorial on higher-order statistics (spectra) in signal processing and system theory: theoretical results and some applications , 1991, Proc. IEEE.

[14]  Kenneth A. Loparo,et al.  Tool wear condition monitoring in drilling operations using hidden Markov models (HMMs) , 2001 .