Off-campus UMass Amherst users: To download dissertations, please use the following link to log into our proxy server with your UMass Amherst user name and password.
Non-UMass Amherst users, please click the view more button below to purchase a copy of this dissertation from Proquest.
(Some titles may also be available free of charge in our Open Access Dissertation Collection, so please check there first.)
Development of a self-diagnostic rolling element bearing
The objective of this work is the development of a rolling element bearing with an integrated sensor for on-line health diagnosis. The work was divided into four distinct phases. ^ The first phase of the development was the structural design of the sensor-integrated bearing. The sensor module was embedded in the bearing via a modification of the outer ring. A methodology was developed for designing the outer ring modification by determining how the modification affected the structural integrity of the bearing. Based on the results of several structural analyses, a design template was created. ^ Dynamic modeling was the second phase of the development. The vibration caused by a defective bearing component was simulated using physical models of the bearing structure. It was found that multiple defects could be grouped into families that produced similar vibration spectra. The effect of a misalignment between the inner and outer rings was also investigated. This type of fault was found to cause predictable load variations on the rolling elements. ^ With the first two phases completed, the third phase of this work was the development of an algorithm for processing the sensor signal. A wavelet-based signal processing algorithm was created to analyze the sensor signal in the time-scale domain This algorithm used a new wavelet which was constructed using an actual impulse response taken from a bearing with two embedded sensor modules. The signal decomposition based on this new wavelet was found to provide a better picture of important diagnostic information contained in the sensor signal. In addition to the wavelet analysis, an adaptive neuro-fuzzy inference system was used to characterize the sensor signal in terms of fault signatures. ^ The fourth phase of this work was to experimentally verify the results obtained in the first three phases. Deep groove ball bearings were used for the experiments, and the tests were conducted at many different loads, speeds, and combinations thereof. Predicted load variations and vibration spectra were observed in the experimental data, which validated the results obtained in the first three investigative phases of this work. ^ The dynamics and kinematics of rolling element bearings were studied, and a time-scale signal processing algorithm was developed in this work. The major contributions are: (1) Development of a novel design for a rolling element bearing with an integrated sensor module. (2) Establishing a means for analyzing multiple bearing defects and grouping them into families that produce similar vibration spectra. (3) Development of a model to predict the output of a load sensor embedded in the outer ring of a bearing. (4) Development of a theoretical framework to generate a new wavelet for analyzing a dynamic sensor signal. (5) Experimental verification of the load carrying and diagnostic capabilities of a bearing with an integrated sensor. ^
Engineering, Electronics and Electrical|Engineering, Mechanical
Brian Thomas Holm-Hansen,
"Development of a self-diagnostic rolling element bearing"
(January 1, 1999).
Electronic Doctoral Dissertations for UMass Amherst.