Research And Verification Of Geared Turbofan Engine Modelling And Disturbance Rejection Control

Geared Turbofan(GTF)engine has a wide market application as one of research directions of the future big bypass ratio civil engine.Large commercial engines will encounter various types of unpredictable disturbance during long term operation in a wide range.These disturbance phenomena pose great challenges to engine stability,aircraft safety,and passenger comfort.So it is very important to study the disturbance problem and disturbance rejection control of the engine.In this paper,the research on engine component level modeling,disturbance modeling in the engine and disturbance rejection algorithm are carried out for the geared turbofan engine.The integration and verification of the full digital simulation platform and hardware-in-the-loop simulation platform are carried out.The specific research content is as follows:First,the component level model and the linear state-space model of the GTF engine are established.According to the Model Based Design(MBD)method,the component module and simulation framework of the T-MATS toolbox are used to establish the component level model of the GTF engine which is above the idle state.The rationality and accuracy of the component level is verified by comparing the data at the non-design point.The linear state space model of the geared turbofan engine is established by the small perturbation and the fitting method.The simulation results show that the error between the component-level model and the linear model is small,which provides a good basis for the design of the disturbance rejection controller.Secondly,analyzing and defining the disturbance phenomena which are existed in the aero-engine,and studying the effects of various disturbance when the system is operating in the open loop.Aiming at the atmospheric turbulence disturbance,a time-domain model of atmospheric turbulence was established by using the kolmogorov spectrum modification method proposed by kopasakis.Aiming at the disturbance of power extraction,a power extraction model referring to the structure of the hybrid engine is established based on the principle of the motor and lithium battery.Aiming at the bleed air disturbance,the pneumatic system model is established according to the bleed air mechanism.The mechanisms of various types of disturbance and their effects on the engine are studied separately.The established disturbance model has high credibility and strong versatility,which is convenient for further research and improvement of the disturbance model in the future.Thirdly,the anti-disturbance principle of H_?control theory and Active Disturbance Rejection Control(ADRC)theory anti-jamming principle are analyzed,and then the LADRC,PI, H_?closed-loop speed controller are designed and compared.Adjusting various controller parameters to make sure that they all approximately have the same rise time,signal tracking ability and control energy when facing the unit step instructions.Then the anti-disturbance effects of the three controllers which are under the different state points and the different disturbance factors are compared.The results show that the active control methods such as LADRC have better anti-interference performance than the traditional passive control methods such as PI and H_?,which ensure the smooth and safe flight of the engine.Finally,the integration and verification of the fully digital simulation platform and the hardware-in-the-loop simulation platform for the models and control algorithms are implemented.The model and controller designed based on the MBD method are integrated into the FWorks simulation platform for testing,which verifying the rationality and accuracy of the model codes and controller codes,and providing guarantee for the integration of the Hardware in the loop(HIL)platform.Through the preparation based on the standard design process,the HIL simulation of LADRC and other algorithms and the disturbance test of GTF engine are finally realized.Simulation results show that active control algorithms such as LADRC have better anti-disturbance performance than ordinary passive control in real controllers.