ABSTRACTS.

This paper presents an innovative prognostic model based on health state probability estimation embedded in the closed loop diagnostic and prognostic system to provide comprehensive timely analysis for effective decision making in industrial asset management. To apply an appropriate classifier in health state probability estimation for prediction of machine remnant life, a comparative study on intelligent fault classification for four fault conditions with five different fault progressed data from high pressure cryogenic pumps were conducted. Five different classifiers, such as support vector machines (SVMs), radial basis function neural networks, random forest and linear regression, were used to evaluate their effectiveness for the health state estimation process. Two sets of impeller-rub data were employed for the prediction of pump remnant life based on estimation of discrete health state probability with feature selection technique using the outstanding capability of SVM. The results obtained were very encouraging and showed that the proposed prognosis system has the potential to be used as an estimation tool for machine remnant life prediction in real life industrial applications. [ABSTRACT FROM AUTHOR]