Analysis techniques of condition monitoring applied to a ball bearing rig (seta)

Osuagwu, Charles Chukwudi (1978) Analysis techniques of condition monitoring applied to a ball bearing rig (seta). University of Southampton, Doctoral Thesis.

Abstract

The assessment of the internal state of a machine from externally acquired control parameters, like vibration and lubrication debris, consists of two phases, fault detection and diagnosis. The advantage of this procedure, called condition monitoring, is that defects can be detected and diagnosed in the incipient stage without machine strip. This results in increased availability, improved safety and economic operation of the machine. The approach used in this thesis for the condition monitoring of a Seta mechanism was to extract discriminant features based on the following dissimilarities between the good and defect acceleration waveformss1. the dual envelope structure of the defect waveform and the randomness of the good waveform; 2. the change in the underlying statistical distribution of the data from normal to non-normal with onset of defects3. the quasi-periodic dependence of the defective waveform and the impact independence of the good waveform.Also,the uniqueness of the impact period of the different Seta components in the incipient phase of defect was utilized for diagnosis. In order to enhance the automatic detection of these periodicities, preprocessing techniques were used to remove the signal transmission path effects from the diagnostic signal. The results obtained have shown that incipient defects in the Seta mechanism can be detected and diagnosed using a number of techniques.

This record has no associated files available for download.

Contributors

Download statistics