Health Monitoring Of Aero-engine Gas Path System Based On Data-driven

Aero-engine is a kind of complex thermal power machinery.With the increasing demand for aircraft performance,the complexity of aero-engine also increases.In order to ensure that the aero-engine can still play its work efficiency in all kinds of harsh environment,the research on the health monitoring technology of its gas system is particularly important.For this reason,this paper uses a variety of methods to study the engine gas path health monitoring technology.The main contents of this paper are as follows:Firstly,the method of selecting and confirming the effective parameters of engine gas path health monitoring based on GCA-PCA is proposed.In the selection and confirmation of the effective parameters,GCA method is used to analyze the correlation between the original parameters of the engine gas path system,PCA method is used to calculate the contribution rate of the original parameters data to its health state,and comprehensive analysis and comparison are carried out to finally determine the effective parameters of the engine gas path health monitoring,so as to complete the selection and comparison of the effective parameters of the engine gas path health monitoring Confirm.Secondly,ICA method is used to extract the data features of effective monitoring parameters of engine gas path.ICA method is used to process the effective monitoring parameter data of engine gas path,and extract the feature information which can better reflect the different states of engine,which can provide effective data for the reliability of the subsequent health monitoring data.Then,the engine gas path health monitoring model of GRNN and DBN is designed and established with effective data.The monitoring effect of the health monitoring model is verified.The results show that the accuracy of GRNN health monitoring model is 87%,while that of DBN health monitoring model is 94%.Finally,in order to further improve the accuracy of health monitoring,based on the above research,this paper also takes the fuzzy integral method to fuse the health monitoring results of GRNN and DBN.The results show that the accuracy of the engine gas path health monitoring model based on fuzzy integral fusion is 96.5%,which is significantly better than the accuracy of GRNN and DBN monitoring models before fusion,so it has a better engineering application prospect.