Decomposition of Symptom Observation Matrix and its Optimization in Vibration Condition Monitoring of Machines


Article Preview

With the modern metrology we can measure almost all variables in the phenomenon field of a working machine, and many of measuring quantities can be symptoms of machine condition. On this basis we can form the symptom observation matrix (SOM) intended for condition monitoring. On the other hand we know, that contemporary complex machines may have many modes of failure, so called faults. The paper presents a method for the extraction of fault information from the symptom observation matrix by means of singular value decomposition (SVD) in the form of generalized fault symptoms. As the readings of the symptoms can be unstable, the moving average of the SOM was applied with success. The attempt to assess the diagnostic contribution of primary symptom was undertaken, and also some approach to connect SVD methodology with neural nets is considered. These possibilities are illustrated in the paper by processing data taken directly from the vibration condition monitoring of the machine.



Edited by:

Tadeusz Uhl and Andrzej Chudzikiewicz




C. Cempel, "Decomposition of Symptom Observation Matrix and its Optimization in Vibration Condition Monitoring of Machines", Applied Mechanics and Materials, Vol. 9, pp. 51-60, 2008

Online since:

October 2007





[1] C. Cempel, Innovative developments in systems condition monitoring, Keynote Lecture, Proceedings of DAMAS'99, Dublin (1999).

[2] C. Cempel, H. G. Natke, J. P. T. Yao, Symptom Reliability and Hazard for Systems Condition Monitoring, Mechanical Systems and Signal Processing, Vol. 14, No 3, 2000, pp.495-505.


[3] C. Cempel, M. Tabaszewski, Multidimensional vibration condition monitoring of nonstationary systems in operation, Proceedings of 12 International Congress on Sound and Vibration, (on CD- paper No 496), Lisbon, July 2005, (full version of the paper sent to Mechanical Systems and Signal Processing - in print). 4. C. Cempel, M. Krakowiak, Influence of running stability and randomness of observation on the condition assessment in multidimensional machine diagnostics, (in Polish), 33 Conference on Machine Diagnostics, Węgierska Górka, (2006).


[5] J. Korbicz, et al., (eds. ), Fault Diagnosis - Models, Artificial Intelligence, Applications, Springer Verlag, Berlin - Heidelberg, (2004).

[6] I. Y. Tumer, E. M. Huff, Principal component analysis of tri-axial vibration data from helicopter transmission, 56th Meeting of the Society of Machine Failure Prevention Technology, (2002).

[7] W. Bartelmus, Z. Zimroz, H. Batra, Gearbox vibration signal preprocessing and input values choice for neural network training, AI Methods Conference, Gliwice, Poland, (2003).

[8] M. Jasiński, Empirical models in gearbox diagnostics (in Polish), PhD Thesis, Warsaw University of Technology, Warsaw, December (2004).

[9] N. H. Pantopian, J. Larsen, Unsupervised condition detection in large diesel engines, Proceedings of the IEEE Workshop on Neural Networks, (1999).

[10] C. Cempel, Implementing Multidimensional Inference Capability in Vibration Condition Monitoring, Proceedings of Conference: Acoustical and Vibratory Surveillance, Senlis - France, October (2004).

[11] H. G. Natke, C. Cempel, The symptom observation matrix for monitoring and diagnosis, Journal of Sound and Vibration, 248, pp.597-620, (2002).


[12] H. G. Natke, C. Cempel, Model Aided Diagnosis of Mechanical Systems, Springer Verlag, Berlin, 1997, p.248.

[13] Golub G. H., VanLoan C. F., Matrix Computation, Oxford, North Oxford Academic, (1983).

[14] Kiełbasiński A., Schwietlick H., Numeryczna Algebra Liniowa, WNT, Warszawa, 1992, (Numeric Linear Algebra, in Polish).

[15] C. Cempel, Multidimensional Condition Monitoring of Mechanical Systems in Operation, Mechanical Systems and Signal Processing, 2003, 17(6), pp.1291-1303.


[16] B. Żółtowski, C. Cempel, (eds. ), Engineering of Machine Diagnostics, (in Polish), ITE Press, Radom, (2004).

[17] C. Cempel, M. Tabaszewski, Averaging the Symptoms in Multidimensional Condition Monitoring for Machines in Nonstationary Operation, (paper sent to) 13 International Congress on Sound and Vibration, Vienna, July (2006).

[18] T. Will, Hanger matrix, two-thirds theorem, Internet; http: /www. uwlax. edu/faculty/will/svd/svd/index. html , June 2005; (see also there: SVD ingredients Mathematica, April 2004).

[19] C. Cempel, Multi fault vibrational diagnostics of critical machines, Zagadnienia Eksploatacji Maszyn, No 4, (144), 2005, pp.133-142.

[20] M. H. Dunham, Data Mining - Introductory and Advanced Topics, Prentice Hall, New Jersey, (2003).

[21] C. Cempel, Vibroacoustic Condition Monitoring, Ellis Horwood, London, (1991).