University of Twente Student Theses
Comparative Study of Fault Detection and Diagnosis for Low-Speed Ball Bearings
Veltman, Mike (2023) Comparative Study of Fault Detection and Diagnosis for Low-Speed Ball Bearings.
|
PDF
2MB |
| Abstract: | Fault Detection and Diagnosis (FDD) in Low-Speed Ball Bearings (LSBBs) is vital for ensuring the reliability and performance of radar systems. This comparative study examines approaches for FDD of LSBBs by collecting vibration and acoustic emission data using sensors on a test bench containing artificially induced bearing defects. Collected data was processed using pre-processing and feature extraction methods, the performance of which was evaluated using Random Forests and Principal Component Analysis. Results indicate that vibration sensing, at a 40 Hz sample rate with one sensor, conveys more information about LSBB defects than acoustic emissions measured between 100 and 450 kHz. Using vibration analysis, defects were successfully detected and identified. While all tested pre-processing methods performed comparably, root mean square and peak frequency magnitude were found most informative for feature extraction on bearing defects. The best-performing combination of methods was matched filter pre-processing combined with root mean square feature extraction. Using lower sample rates and fewer vibration sensors offers potential cost savings and increased computational efficiency. Important to consider is the use of test bench data with artificial defects, which may not fully represent real-world radar systems. Therefore, data from real-world systems and bearing statuses is necessary to confirm the generalizability of the recommended approach. Despite these limitations, the findings provide valuable insights into FDD for LSBBs in radar systems, as well as in similar systems. These insights contribute to the improvement of radar system maintenance and reliability and ultimately to a safer, more secure and more efficient maritime environment. |
| Item Type: | Essay (Master) |
| Faculty: | EEMCS: Electrical Engineering, Mathematics and Computer Science |
| Subject: | 50 technical science in general, 54 computer science |
| Programme: | Embedded Systems MSc (60331) |
| Link to this item: | https://purl.utwente.nl/essays/95018 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page