Fault Diagnosis of Civil Aircraft Avionics System Based on Bayesian Network and Function Ground Test

Abstract

A fault diagnosis method based on Bayesian network is designed for avionics system in this paper. This method can automatically diagnose possible faults after system assembly through function ground test results. Based on the function ground test, this method establishes a Bayesian network consisting of four kinds of nodes: function test node, function fault node, system fault node and minimum component node. Afterwards, taking the test pass rate and fault occurrence rate of the function test node as the prior probability of the network and passing through the function fault node and the system fault node, the minimum component node fault probability is diagnosed through network reasoning. Combined with the actual test data, the experimental results show that the diagnosis results are in line with the actual situation, and verify the effectiveness of the model. The experimental results provide an effective basis for fault diagnosis of avionics system and safety management of civil aircraft system assembly.