Aiming at the fault diagnosis for nonlinear

non-Gaussian systems monitored by multiple sensors

a novel particle filtering based approach to distributed fault diagnosis is developed.One of its advantages is that the complete probability distribution information of state estimation from particle filtering is utilized for fault detection.Firstly the distributed fault model of system is set up

and it is assumed that sensors can only send binary data to the fusion center because of the communication constraints.Under the assumption of independent

identically distributed observations

a distributed fault detection algorithm is proposed

including local detector design and decision fusion rule design.Simulation results show the efficiency and superiority of our proposed algorithm.