HMM Based Modeling and Health Condition Assessment for Degradation Process

Lisha Xia,Huajing Fang,Hui Zhang
DOI: https://doi.org/10.1109/ccdc.2013.6561449
2013-01-01
Abstract:The modeling and health condition assessment for degradation process are crucial to the effective machine fault diagnosis and prognosis. They provide a potent tool for operators in decision-making by specifying the present machine state and estimating the remaining useful life (RUL). In this paper, the health conditions of degradation process are modeled as a hidden Markov chain and the physical outputs are modeled as the stochastic events whose probability depends on the Markov chain state. The expectation maximization (E-M) algorithm is proposed to learn parameters of the modeled hidden Markov model (HMM) and the iteration convergence is demonstrated. A maximum a posteriori (MAP) current health condition assessment approach is also proposed.
What problem does this paper attempt to address?