Font Size: a A A

Research Of Fuzzy Fault Detection And Diagnosis Methods For Liquid-Propellant Rocket Engines

Posted on:2009-05-30Degree:MasterType:Thesis
Country:ChinaCandidate:Y J LiFull Text:PDF
GTID:2132360278957047Subject:Aeronautical and Astronautical Science and Technology
Abstract/Summary:PDF Full Text Request
This thesis puts an emphasis on the selection of monitoring parameters, extracting of fault characteristics, the modeling and identification of operational process and the real-time fault detection and isolation for a certain large turbo-pump fed liquid-propellant rocket engine (LRE).The real-time fault detection algorithms based on fuzzy model for a start-up process and main stage process of rocket engine are designed and realized.Firstly, the status of research of health monitoring system and fault diagnosis methods at home and abroad, is introduced, then the fault modes in an operating process of rocket engines are analyzed, and the parameters which needed to be monitored during the operation of engine are selected by the methods RC(regularity criterion).Secondly, the basic principles of fuzzy model are briefly introduced, then the application of T-S(takagi-sugeno) fuzzy model into the fault detection of LRE is particularly analyzed; and pretreated the training data with FCM(Fuzzy c-means clustering) method for improving on the training of the fuzzy model.Then modeling and identification of operational process of the LRE start-up and main stage, combined with the fault detection logic based on fuzzy logic, the fault detection algorithms for a start-up process and main-stage operating process of LRE are realized. The fuzzy algorithms isolate the fault at main stage primary.The fault detection algorithms based on fuzzy model are off-line validated by a large amount of fire-test data. The results show that the fault detection algorithms based on fuzzy model can detect the faults timely and effectively which have occurred during the test with no false alarm to normal test data. The algorithms can meet the requirements of real-time ability and robustness during the ground fire-test, it also can isolate the main stage fault effective. The results have some important significance in the application to develop the health monitoring system for LRE and launch vehicles.
Keywords/Search Tags:Liquid-propellant Rocket Engine, Model Identification, T-S(takagi- sugeno) fuzzy model, Fault Detection and Isolation
PDF Full Text Request
Related items