Where Next for Condition Monitoring of Rotating Machinery?

MI Friswell (University of Bristol) & AW Lees (University of Wales Swansea)

Advances in Vibration Engineering, Vol. 5, No. 4, 2006, pp. 263-277

Abstract

Condition monitoring has progressed for many years with significant but fairly incremental improvements to algorithms for data analysis. Pattern recognition methods for the condition monitoring of rotating machinery are well established and can successfully detect and classify many types of faults. However some faults are difficult to distinguish and locating faults requires a spatial model of the machine. To be useful, these models must be an accurate representation of the machine. Of course, as with any model, the required accuracy depends on the use to which the model is put. This paper considers models of various faults and machines, highlights the use of estimation methods, and discusses the required model fidelity. Examples of faults such as cracks, unbalance, bends and misalignment are given. Consideration is given to the information that may be established about an operational machine, and how an operational system might be implemented in future condition monitoring systems to significantly enhance their value.

Paper Availability

This material has been published in the Advances in Vibration Engineering, Vol. 5, No. 4, 2006, pp. 263-277. Unfortunately the copyright agreement with The Vibration Institute of India does not allow for the PDF file of the paper to be available on this website.


Advances in Vibration Engineering