Bearing wear prognosis based on Hidden Markov Model

Abstract

Condition-based maintenance (CBM) is a maintenance program that recommends maintenance decisions based on the information collected through condition monitoring. There are two substantial factors to support CBM, machine fault diagnosis and machine condition prognosis. In this paper deal with an estimation of the Remaining Useful Life of bearing based on the Hidden Markov Models. The Prognostic process is done in two phase: a learning phase and evaluation phase. In first process, the sensors’ data are processed in order to extract appropriate features, which are used as inputs of learning HMM. During second phase the extracted features are continuously injected to the obtained model to represents the current health state of mechanism system and to estimate its remaining useful life. The proposed method is tested on bearing wearing test of Rotor kit RK4.