基于粒子滤波的自适应转移混合系统故障检测.pdfVIP

130 IEEE TRANSACTIONS ON RELIABILITY, VOL. 61, NO. 1, MARCH 2012 Particle Filtering for the Detection of Fault Onset Time in Hybrid Dynamic Systems With Autonomous Transitions Francesco Cadini, Enrico Zio, and Giovanni Peloni Abstract— The behavior of multi-component engineered systems Time step. is typically characterized by transitions among discrete modes of operation and failure, each one giving rise to a specific continuous The total number of time steps in the dynamics of evolution. The detection of the system’s mode change observation period. time represents a particularly challenging task because it requires keeping track of the transitions among the multiple system dy- Number of particles. namics corresponding to the different modes of operation and -th possible hypothesis of the probability failure. To this purpose, we implement a novel particle filtering distribution method within a log-likelihood ratio approach here, specifically tailored to handle hybrid dynamic systems. The proposed method Initial distribution of the system state. relies on the generation of multiple particle swarms for each discrete mode, each originating from the nominal particle swarm Transition probability distribution (prior). at different time instants. The hybrid system considered consists of a hold up