A new bearing defect detection scheme is proposed using the dala dependent systems (DDS) methodology. The detection scheme has two parts: monitoring and diagnosis. Monitoring indicates the stability of the defect and diagnosis identifies when it is significantly large and its location. The detection scheme uses two separate indices: band energy and detection vector. Two life-tests were used to judge the effectiveness of the scheme. In both tests, the scheme identified the defect (on the inner-race and the outer-race) several days before failure occurred. The detection scheme thus provides excellent results when predicting both types of defects.
All Science Journal Classification (ASJC) codes
- Theoretical Computer Science
- Computer Science Applications
- Computational Theory and Mathematics
- Artificial Intelligence