A PROGNOSIS METHOD USING HIDDEN SEMI-MARKOV MODEL FOR EQUIPMENT HEALTH PREDICTION

摘要

Health monitoring and prognostics of equipment is a basic requirement for condition-based maintenance in many application domains where safety, reliability, and availability of the systems are considered mission critical. Conducting successful prognosis, however, is more difficult than conducting fault diagnosis. Prognosis generally requires a sound understanding of asset condition history. A much broader range of asset health related data, especially those related to the failures, shall be collected. The asset health progression can then be possibly extracted from the congregated data, which has proved to be very challenging.The hazard function (h.f.) has been used to analyze the distribution of lifetime with a combination of historical failure data and on-line condition monitoring data. Using h.f., CBM is based on a failure rate which is a function of both the equipment age and the equipment conditions. The state values of the equipment condition considered in CBM, however, are limited to those stochastically increasing over time and those having non-decreasing effect on the hazard rate.This paper presents a hidden semi-Markov model (HSMM) based prognosis method for prediction of equipment health. A HSMM is a hidden Markov model with temporal structures. Unlike standard hidden Markov model (HMM), HSMM does not follow the unrealistic Markov chain assumption and therefore provides more powerful modeling and analysis capability for real problems. In addition, an HSMM allows modeling the time duration of the hidden states and therefore is capable of prognosis. The estimated state duration probability distributions can be used to predict the remaining useful life of the systems. The previous HSMM based prognosis algorithm assumed that the transition probabilities are only state-dependent. That is, the probability of making transition to a less healthy state does not increase with the age. In the proposed method, in order to characterize a deteriorating machine, an aging factor that discounts the probabilities of staying at current state while increasing the probabilities of transitions to less healthy states will be introduced. With the equipment health prognosis, we can predict the behavior of the equipment condition.