Extraction of Bearing Fault Transients from a Strong Continuous Signal Via DWPA Multiple Hand-Pass Filtering

Abstract

This paper presents a new method to enhance the detection and diagnosis of rolling element-bearing faults based on discrete wavelet packet analysis (DWPA). The extraction of attenuated resonant vibrations due to impacts from localized faults in rolling element bearings is normally achieved by high-pass or band-pass filtering of the vibration signal. The main problem with this approach is the difficulty in choosing an appropriate filter range of interest. This is a serious obstacle when the bearing fault transients are buried in high levels of noise or contaminating signals. An alternative that enables the automation of the selection process and the inclusion of multiple frequency bands of interest is presented. A superior signal to noise ratio is achieved in comparison to either high-pass or band-pass filtering of the signal, as the DWPA feature extraction facilitates the equivalent of automatically selecting an optimal multiple band-pass filter.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Apr 05, 2001
Accession Number
ADP013502

Entities

People

  • Joseph P. Mathew
  • Joshua Altmann

Tags

Communities of Interest

  • Energy and Power Technologies

DTIC Thesaurus Topics

  • Algorithms
  • Amplitude
  • Carrier Waves
  • Contamination
  • Data Sets
  • Extraction
  • Feature Extraction
  • Filters
  • Filtration
  • Frequency
  • Harmonics
  • Least Squares Method
  • Mathematical Models
  • Medical Engineering
  • Models
  • Reliability
  • Spectra

Fields of Study

  • Engineering

Readers

  • Computer Vision.
  • Fault Tolerant Diagnosis of Black and White Balloon Isolation Tests Using ¥.
  • Radar Systems Engineering.

Technology Areas

  • AI & ML
  • AI & ML - Machine Learning Algorithms
  • AI & ML - Neural Networks