Abstract
Abstract
Condition monitoring of machines with data sampled from time-dependent variations has been elucidated. Passive diagnostic approaches have been examined particularly for prognostic parameters. Trend predictions using truncated power series and autoregressive moving average methods have been compared. Necessary background information for modelling with respect to measured data acquired over a period of 2 years on a specific case has been provided. The repair effected on the basis of trend forecasting exemplifies an important application of condition monitoring.
Keywords
Get full access to this article
View all access options for this article.
