Font Size: a A A

Research On Ypical Fault Monitoring Method Of Aero Engine

Posted on:2018-11-07Degree:MasterType:Thesis
Country:ChinaCandidate:H W LiuFull Text:PDF
GTID:2322330512973272Subject:Electronic and communication engineering
Abstract/Summary:
The main fuel pump of aeroengine as one of the core components of the aircraft engine fuel system,whether it can work normally will directly affect the aircraft flight safety.As the main fuel pump work in a long period of high pressure,high temperature and other harsh working environment,therefore,the main fuel pump is prone to failure and the general life is short,how to accurately identify the key components of the aircraft engine in which different operating conditions,to ensure the safety of aircraft flight and reduce maintenance costs play a key role,it has important military and economic value.In this paper,the key component of a certain type of aeroengine-the main fuel pump for the specific object of study,to have a research on the state monitoring technology of main fuel pump in the course of flight encountered in the typical failure.Firstly,on the test platform of a certain type of real aeroengine,the original data of many parameters that characterize the running state of the aeroengine are obtained through a long-term test.Research on the method of data validity of the main fuel pump after the data pretreatment,to obtain the characteristic parameters which can characterize the health status of the main fuel pump.After comprehensive analysis,the health state of the main fuel pump is divided into four types,including normal state,the main fuel pump bearing fault,the main fuel pump regulator failure,and the main fuel pump exhaust temperature and speed exceeds the limit value of the phenomenon when the throttle lever from the local train pushed to the middle state aviation.Finally,three methods of the fault diagnosis model are built,including SVM,ELM and KELM.The fault diagnosis technology of the main fuel pump is studied by using the constructed model.The research results show that the SVM fault diagnosis rate is 87.5%,the ELM fault diagnosis rate is 92.5%,the KELM fault diagnosis rate is 97.5%.Therefore,the condition monitoring algorithm designed in this paper can be used for failure mode of the main fuel pump,which has reached the expected target.
Keywords/Search Tags:The main fuel pump, Principal component analysis, SVM, ELM, KELM
Related items