Gas-path Fault Diagnosis Of Turbofan Engine Based On Decision Information Fusion

The fault diagnosis technology of aero-engine gas path components is the key to realize system maintenance,reduce flight cost and guarantee system reliability.It is of great significance to study it.This thesis mainly studies the fault diagnosis methods of turbofan engine based on data-driven technology and information fusion technology.The main contents of the paper are as follows:First,we state the research background and significance of the topic,introduce the state-of-art both foreign and domestic development of theoretical and practical application of engine fault diagnosis research,also we enumerate some methods of gas path fault diagnosis,and analyzes the categories,causes and characteristics of fault.Second,a fault diagnosis system based on information fusion technology for turbofan engine gas path components is designed.In view of the shortcomings of neural network algorithm,for its low accuracy,the training speed is slow low training speed,difficulties in determining the network nodes trapped in local extremum slow convergence speed.This thesis overcomes the shortcomings of the BP neural network mentioned above by introducing momentum factor.The improved BP neural network and support vector machine are then used to make a preliminary diagnosis of the failure turbofan engine components failure,and the diagnosis results are used as the evidence body for decisionmaking diagnosis.At last,Based on measurement parameters and fault characteristics of turbofan engine,a identification framework is constructed to determine the failure symptoms.Combined with the characteristics of improved BP neural network and support vector machine,a new basic probability function allocation method is proposed and decision rules are formulated.Diagnostic research on turbofan engine gas path faults at the decision-making level.Finally,the information fusion diagnosis algorithm is tested for anti-interference.The test verifies the correctness of the information fusion diagnosis algorithm and provides a reliable and accurate diagnosis method for gas path fault diagnosis.The results show that the proposed methods can meet the needs of the actual diagnostic tasks of turbofan engines.