Algorithms for statistical moment evaluation for machine health monitoring

✍ Scribed by H.R. Martin; F. Ismail; A. Sakuta

• Publisher: Elsevier Science
• Year: 1992
• Tongue: English
• Weight: 471 KB
• Volume: 6
• Category: Article
• ISSN: 0888-3270
• DOI: 10.1016/0888-3270(92)90034-g

✦ Synopsis

The analysis of vibration signals has proven to be a powerful and effective tool for the early detection of developing failure mechanisms in machines. As an alternative approach to the traditional spectrum methods, analysis can be carded out on time domain signals directly using statistical methods. This is particularly useful where a high content of random signal is present. This paper presents algorithms that can be used to estimate statistical moments of both Gaussian and beta distributions, for these applications. It is demonstrated here that fitting the raw data on the beta distribution reduces the sensitivity to noise and thus guards against false alarms.