Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


31802.pdf (6.78 MB)

ETD Abstract Container

Abstract Header

Design of a Classifier for Bearing Health Prognostics using Time Series Data

Iyer, Balaji S
http://rave.ohiolink.edu/etdc/view?acc_num=ucin1543922781446885

Abstract Details

2018, MS, University of Cincinnati, Engineering and Applied Science: Electrical Engineering.
Accurate prediction of residual life of bearings is a challenging problem in machine prognostics. Instead of treating this problem as one of regression, which is the case with most literature in this field, this thesis focuses on developing a binary classifier to address this problem. The decision of the classifier corresponds to whether the bearing set is likely to fail within the next 3 weeks or not given vibration data of the previous 10 hours. Critical periods in the bearings life, where the bearing behavior changes significantly compared to previously observed behaviors, are identified using the Bhattacharyya distance. Using k-nearest modes, a strategy similar in spirit to k-nearest neighbor classification, the bearings are classified. Furthermore, a range compensation algorithm is proposed for utilizing modes observed in one set of bearings to classify vibration data from a different set of bearings. This significantly enhances the applicability of the algorithm. The algorithms are tested using real world vibration data collected from a spindle test-bed built by TechSolve Inc. The proposed methodology performs favorably over a large range of parameters when compared with a Gaussian kernel SVM.
Raj Bhatnagar, Ph.D. (Committee Chair)
Gowtham Atluri, Ph.D. (Committee Member)
Yizong Cheng, Ph.D. (Committee Member)
Radu Pavel, Ph.D. (Committee Member)
83 p.
Electrical Engineering
Bearing Health; Time Series; Classifier; Bhattacharyya Distance; Classifier; Power Spectral Density

Recommended Citations

Citations

  • Iyer, B. S. (2018). Design of a Classifier for Bearing Health Prognostics using Time Series Data [Master's thesis, University of Cincinnati]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1543922781446885

    APA Style (7th edition)

  • Iyer, Balaji. Design of a Classifier for Bearing Health Prognostics using Time Series Data. 2018. University of Cincinnati, Master's thesis. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=ucin1543922781446885.

    MLA Style (8th edition)

  • Iyer, Balaji. "Design of a Classifier for Bearing Health Prognostics using Time Series Data." Master's thesis, University of Cincinnati, 2018. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1543922781446885

    Chicago Manual of Style (17th edition)

ucin1543922781446885
199
© 2018, all rights reserved.
This open access ETD is published by University of Cincinnati and OhioLINK.