Fault Diagnosis Method for Avionics System based on Conditional Fuzzy Petri Nets

In order to analyse the fault of avionics system effectively, the conditional places are added on the basis of traditional fuzzy Petri nets. Firstly, by analysing the functional structure of the system and combining with the maintenance data, the forward reasoning model is established by using conditional fuzzy Petri nets for fault propagation analysis. Based on the forward reasoning model, the probability formula is combined to construct the backward reasoning model, and the certainty factor (CF) of partial reverse transitions is timely updated according to the state of the conditional places for fault diagnosis. Then, given the initial CF of the places, the forward and reverse reasoning model are quantitatively calculated by using the iterative algorithm based on maximal algebra, and the calculated result of each place in the next state is deduced to analyse the possibility of failure. Finally, an example is given to verify the effectiveness of the method.