Research Of Fault Diagnosis Of Automatic Flight Control System Based On FPN

With the rapid development of civil aviation industry in recent years, the reliability of civil aircraft has become increasingly demanding. The fault detection,isolation and exclusion of aircraft 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 airlines, so how to make maintenance of civil aircraft fast and correct is critical. Considering the fault features of aircraft autopilot in B737 Automatic Flight System(AFS), and using the capabilities and efficient analysis of fuzzy Petri net(FPN) in dealing with complex systems modeling,we study and develop the fault diagnosis expert system based on FPN for B737 AFS in order to improve the efficiency of fault diagnosis.This thesis firstly introduces the knowledge representation and reasoning mechanism of FPN. It focuses on the improvement of FPN concurrent reasoning mechanism and this transformation algotithm not only avoids unnecessary repetitions of firing of transitions effectively, but also improves the efficiency of the reasoning process. Secondly, the artical introduces the intelligent fault diagnosis models of autopilot, and the simulation proves that the improved algorithm is superior in terms of reasoning speed. Then the thesis introduces the establishment of the expert system, and the article diagnoses the faults of autopilot with the combination of the expert system and the Petri net theory, and emphasizes on the research about the fuzzy Petri net introduced into expert system as a tool for knowledge represention and reasoning. Finally, according to the object-oriented technology, a practical fault instance of autopilot is analyzed, which is developed by Visual C++6.0 and SQL Sever 6 softwares, to demonstrate the feasibility and validity of the expert system combined with FPN. The system not only has strong self-learning ability, but also has fast and accurate characteristics.