Rotating Machine Vibration Analysis using Group of Adaptive Models Evolution. Adam Docekal, Marcel Kreidl, Radislav Smid. IWIM, Prague, 2007.
Article (in pdf)
Abstract. Modern rotating machinery often takes advantage of new designs of used gearwheels and rolling bearings. Usage of these new components enables machine to work quieter, increase its reliability, and lengthen working life. Machine vibration analysis belongs to important methods used for rotating machine conditions monitoring. Rotating machine vibrations are often processed by Fourier spectrum especially Power Spectrum Density (PSD) analysis. Power Spectrum Density of vibrations measured on rotating machine contains a lot of peaks at many different frequencies. Recognition of peaks produced by an analyzed part of the machine is necessary for machine monitoring. This could be accomplished by theoretical estimation of frequencies characteristic for the analyzed part, e.g. tooth frequency for gear transmission. These theoretical estimations are unknown for new designs of a selected part or even these estimations are often uncertain. Frequencies or bands important for the solved task have to be estimated directly from the measured data too. Group of Adaptive Models Evolution was used for this purpose for tapered roller bearing analysis and gear transmission vibration analysis.
Keywords. Group of Adaptive Models Evolution, vibration analysis, spectra, important bands
References.
- Dynybyl V., Moss´oczy P., ˇ Ceˇsp´ıro Z.: Experimental Assessment of Bevel Gearboxes. Proceedings of 22nd Danubia-Adria Symposium on Experimental Methods in Solid Mechanics, Parma, University of Parma, 2005, p. 296 – 297.
- Dynybyl V., Doˇcekal A., ˇZ ´ak P.: Experimental research of gearwheels behaviour. Proceedings of Acta Mechanica Slovaca, 2006, vol. 10, no. 4–B/2006, p. 119 – 124, ISSN 1335–2393.
- Kreidl M. et al: Diagnostic Systems. Czech Technical University Publishing House, Prague, 2001, 352 p., ISBN 80–01–02349–4.
- Randall R. B., Antoni J., Chobsaard S.: The Relationship between Spectral Correlation and Envelope Analysis in the Diagnostics of Bearing Faults and Other Cyclostationary Machine Signals. Mechanical Systems and Signal Processing 15(5), p. 945 – 962, 2001.
- Kord´ık P., ˇSnorek M.: Progress in inductive modelling. Proceedings of EUROSIM 2007, Paris, 2007.
- Kord´ık P.: Fully Automated Knowledge Extraction using Group of Adaptive Models Evolution. PhD thesis, FEE, CTU Prague, September 2006.
- Kohavi R.: A study of cross-validation and bootstrap for accuracy estimation and model selection. Proceedings of the Fourteenth International Joint Conference on Artificial Intelligence, San Mateo, p. 1137 – 1143, 1995.
- Schr¨oder H., Blume H.: One- and Multidimensional Signal Processing. John Wiley & Sons, Ltd., New York, 2000.
