HMM based modeling and health condition assessment for degradation process

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.