Research On Aero Engine Fault Diagnosis Based On Fault Tree

Aero engine is one of the most important parts of the aircraft, the structure and systems are very complex, its working environment is very poor, so it is prone to failure, and the engine faults are the main cause of airplane accidents. Therefore, in order to ensure the safety of the engine and save repair costs, aero engine fault diagnosis technology research is very necessary. The method of the aero engine fault diagnosis based on fault tree analysis is studied. The rule-based expert system is reviewed and finally the aero engine fault diagnosis expert system is designed and realized. The system can communicate with users through the man-machine interface, and gradually determine the fault location and reason. And then it can draw conclusions and give corresponding diagnostic maintenance recommendations to achieve expert-level aero engine fault diagnosis.Firstly, the principle of fault tree analysis method is studied, focusing on how to establish the fault tree and fault tree qualitative analysis. Secondly, the basic structure of rule-based expert system is reviewed, especially the functions and establish requirements. The author proposes the combination approach with fault tree analysis method. Then, a large number of engine faults are acquired. The technical literature, the expert experience and manuals knowledge are combined in order to analyze the typical turbofan engine failures and CFM56-7B engine oil, air entraining system failures and establish the fault trees. The generative knowledge representation method and frame representation method are combined. The table structure of knowledge base is designed. The diagnosis knowledge base of expert system is realized. Finally, the diagnostic reasoning process is realized, using forward reasoning method to form reasoning machine, and the knowledge utilization is achieved.The diagnostic knowledge base is created by using SQLserver2008 database platform; C# has an outstanding ability of integrated development and friendly visual interface, the aero engine fault diagnosis expert system and the knowledge base management system are written by C#.