Studies On Fault Diagnosis Expert System Of B737 Aircraft Automatic Flight System

The reliability of civil aircraft is extremely required, and the troubleshooting of aircrafts is not only closely relative to the flight safety, but also directly or indirectly has an important influence on operation benefit and social status of aircraft customer, so how to give the correct schemes of airplane line maintenance efficiently within short time is critical to the maintenance of civil aircraft. It is a good idea to utilize the expert system to produce the airplane line maintenance system. Considering the complexity of malfunction models of different aircraft components in B737 Automatic Flight System(AFS),study and develop the fault diagnosis expert system for B737 AFS in order to improve the quality of fault diagnosis.On the basis of AFS analysis, we collected and generalized the valuable experience of the maintenance engineer in the airline company. According to the fault diagnosis records and the FIM, which provided by the airplane manufacturer, made up of the main data source.Establish a fault diagnosis expert system based on the combination of case-based reasoning and decision fault tree inference. Study the key technology of aircraft troubleshooting based on CBR. It includes the case expression, casebank index strategy, case storage, revise and retain of troubleshooting case. And establish the model of case retrieve combining the matching technology of character fields with KNN method and take an example. Propose a method that made the fault isolation procedures in FIM extract into the form of Decision Fault Tree, establish the frame knowledge representation of Augmented Fault Tree and give the rapid diagnosis of strategy. It has finished the design and maintaining of knowledge base, and the realization of reasoning machine. Put forward a plan about online consultation of the airplane line maintenance in order to share the civil aviation expert resource and attain the best maintenance methods.B737 AFS fault diagnosis expert system prototype is developed by means of VB6.0based on SQL Server. Put forward some ideas about how to improve the system.