Health status management of spacecraft power system based on Hidden Markov Model

The anomaly detection and prediction of spacecraft power system is an essential factor in determination of the spacecraft health status. This paper proposed a Hidden Markov Model (HMM)based approach to the problem. Health status set was first established according to expert knowledge and operation experience. Model parameters, health index and baseline were then designed for HMM. Probabilities of health status were finally determined using observed data with the HMM method. Two experiments were conducted to demonstrate the effectiveness of our new approach.