Research On Degradation Aero-engine's Parameters Estimation And Control Technology

In order to ensure the aircraft engine remaining health,safe,reliable and available state in the process of fight,this paper around the problem of engine gas path and sensor failure and performance recovery,we developed study for the degradation aero-engine's parameters estimation and control technology.Furthermore,a digital simulation was carried out.Firstly,this paper introduced the mathematical model of turbofan engine's modeling process,and all of the interrelated research was based on this nonlinear component level model.The various components of engine's degradation mechanism were analyzed and the method of setting degradation parameter was used to indicate the turbofan engine's degradation state.The effect of different component performance degradation on the performance parameters of turbofan engine was simulated,the digital simulation shows that the effect of component deterioration on the engine could be accurately indicated by this model.Secondly,for the problem of the aero-engine gas path sensor fault,this paper designed turbofan engine's gas path sensor fault diagnosis system which was based on online sequential extreme learning machine algorithms.In order to reduce the fault diagnosis time,the selection strategy was adopted which the output layer weight vector was updated selectively based on generalization capability.For the problem of using fault sensor's measurement signal as the input of the algorithms which could lead the diagnosis system chaos and collapse,the fault isolation system was designed which was based on the prediction learning,through the isolation system which can remove the fault sensor signal from the inputs information,it can improve the diagnostic system's robustness.The digital simulation shows that the sensor diagnosis system can fast locate fault sensor and isolate the sensor's signal whether it happened single sensor failure or multiple-sensor failures.Moreover,for the problem of the aero-engine gas path components fault,this paper established different static working point‘s linear component level model in the study of fight envelope,and then combined with the improved Kalman filter,according to current altitude,Mach and fuel flow,used 3d-interpolation method to estimate the current working point's component degradation parameters.According to degradation parameters and the current flight status,MRR-LSSVR algorithm was used to estimate the engine performance parameters such as thrust.In view of the contradiction that actual aero-engine sensors' number less than degradation parameters,the method of SVD-KF and virtual sensor was used to solved this problem.Finally,for the problem that the degradation engine's important parameter-thrust deviated from the rated condition,performance recovery control technology was studied in this paper.The degradation engine's thrust estimators based on MRR-LSSVR was designed,and the augmented LQR was used as the inner loop controller.Furthermore,expert PI controller was designed as the outer loop controller to adaptively control inner loop controller's rotate speed.The simulation results show that the performance recovery control system could recovery the degradation engine's thrust to the rated condition,fully extracting the potential performance of the degenerate engine.