Abstract
Recently, a lot of researches have been conducted in applying Continuous Wavelet Transforms (CWTs)to vibration based machine fault diagnosis. Since CWTs can provide multi-resolution in the time-frequency-amplitude analysis, they are very usefu1 in detecting both periodically and randomly occurred anomalous signals. However, due to the phenomena of coefficient overlapping and energy leakage, the results generated by CWTs are difficult to be interpreted by in experienced operators. Misinterpretation of results could lead to false alarms or failures to detect anomalous signals, Such negligence may cause fata1
breakdowns of machines that interrupt production and services. These deficiencies make CWTs sti11not popular to be used in vibration based machine fault diagnosis. This paper presents the findings from an investigation on the problem of overlapping and introduces a new algorithm to minimize the problem Preliminary results show that the new algorithm is effective in reducing the distortion caused by the overlapping and extracting fault related signals, Hence, with the help of this new algorithm, even in experienced operators can detect the anomalous signals precisely and determine the cause of fault easily.
breakdowns of machines that interrupt production and services. These deficiencies make CWTs sti11not popular to be used in vibration based machine fault diagnosis. This paper presents the findings from an investigation on the problem of overlapping and introduces a new algorithm to minimize the problem Preliminary results show that the new algorithm is effective in reducing the distortion caused by the overlapping and extracting fault related signals, Hence, with the help of this new algorithm, even in experienced operators can detect the anomalous signals precisely and determine the cause of fault easily.
| Original language | English |
|---|---|
| Pages | 9-16 |
| Number of pages | 8 |
| Publication status | Published - Sept 2002 |
| Event | International Symposium on Machine Condition Monitoring and Diagnosis: Mechanical Engineering Congress, Japan 2002 - University of Tokyo, Tokyo, Japan Duration: 25 Sept 2002 → 26 Sept 2002 https://www.jsme.or.jp/2002am/ |
Conference
| Conference | International Symposium on Machine Condition Monitoring and Diagnosis |
|---|---|
| Abbreviated title | JSME annual meetings |
| Place | Japan |
| City | Tokyo |
| Period | 25/09/02 → 26/09/02 |
| Internet address |