Application Research Of Predictive Control In Engine Transition State Control

The engine state can be divided into steady state and transient state according to whether the performance variable changes with time.And the designed require for the transition state's control law is to ensure the realization of good dynamic performance without exceeding the physical or safety limits of the engine.However,because the transition state control is nonlinear in nature,the control effect of the traditional linear controller often fails to meet the requirements when the state changes widely.Therefore,the research on engine transient state control has always been a research hotspot and difficulty in the field of aero-engine control.At present,for the limitation requirements in the transient state control of aero-engine,the commonly used method is to use the method of linear controller and Min-Max selector for limit management.However,the conservative nature of this architecture has restricted the further development of engine potential.In order to solve this problem,the model predictive control algorithm that can cancel the select structure and directly take constraints into consideration is chose and the application of it in the aero-engine transient state control has been studied.The specific research work is as follows:Firstly,a component-level model of a large bypass ducted turbofan engine with split tailpipes and unadjustable exhaust nozzle areas is established.Based on the derivation of the mathematical model of each component,the steady-state and the dynamic models of turbofan engines are established with Matlab/Simulink software.And the guide vane angles and guide angles of related components are considered during the model building process.The characteristics of this engine are analyzed through simulation,and based on this,a linearization method is designed using one-step least square method and fitting method.Then,with the built engine model,a traditional limit management control system is established by using linear controllers and the Min-Max selector,and its conservative problems are exemplified.Aiming at these conservative problems,the model predictive control algorithm that can directly take constraints into consideration is introduced,and the selection structure of traditional methods is eliminated.An aero-engine single-variable controller was designed guided by the MPC algorithm,and its effectiveness was verified by simulation.Then the influence of the controller's key parameters are studied,and the law of parameter selection is revealed.Further,to the limitations of the univariable controller,a multivariable controller based on the MPC algorithm is designed,and the effect of the transition state control under the action of the two controllers is compared and analyzed.Finally,the design method of the wide range model predictive controller is studied.It is mainly divided into two steps.The first step is to extend the control range of the controller from a small speed range to a large speed range by the method of segmented controller and speed scheduling scheme.The main research contents include the study of the segmented controller method,the method of changing the state matrix,and the speed scheduling method.The second step is to replace the flight envelope area with the area feature points by using the flight envelope division method,and use the flight height and flight Mach number scheduling scheme to achieve the wide range controller.Its main research contents include the K-means clustering analysis method,height and Mach number scheduling scheme design and all-envelope controller structure design.The controller designed in this paper is verified by experiments using the nonlinear componentlevel model simulation platform built before.The simulation results show that the engine will not exceed its operating limit range,regardless of the kind of the transient state.Whether the wide range change of low-speed shaft speed or the flight conditions change,its dynamic performance meets design requirements,which verifies the effectiveness of the aero-engine transient state control algorithm designed in this paper.