Research On Multivariable Adaptive Control Algorithm Of Turbofan Engine Based On MRAC

Aero-engine is a complex thermal system with strong nonlinearity.It is difficult to describe it with an accurate model.At the same time,the analysis of the model will be affected by the operating state and flight conditions.The accuracy of the control method designed according to the rated engine model under such conditions is often difficult to get a guarantee.Unlike traditional control methods,the model reference adaptive control method in modern control theory requires low accuracy for the aero-engine model,and can also control the aero-engine when the system parameters change,further excluding other factors and getting expected output.This provides new ideas for the design of aeroengine control systems.In order to make aero-engine overcome the influence of nonlinear characteristics and uncertain factor and operate with the desired performance efficiently and stably in a wide range,this paper chooses model reference adaptive control based on high-frequency gain decomposition to do research on the twin-shaft turbofan engine as follows from different angles under different working conditions:First,the MRAC method of high-frequency gain decomposition of turbofan engine based on linear model is studied.The high-frequency gain is decomposed into three multiplied matrices,and the requirement for the prior knowledge of the engine is only reduced to the positive or negative of sequential principal subunits of the high-frequency gain sequence.The design of the adaptive law is completed in conjunction with the Lyapunov stability theorem.After analyzing the simulation results,in addition to the establishment of the verification algorithm,the influence of adaptive parameters on the engine control results is obtained,which provides theoretical support for the following research.Secondly,a MRAC method for high-frequency gain decomposition of turbofan engines based on the EME model(short for equilibrium manifold expansion model)is proposed.Based on the existing adaptive algorithm,the knowledge of the multi-input multi-output engine EME model is introduced to prove the effectiveness of the MRAC method of high-frequency gain decomposition of the turbofan engine based on the EME model in the range of equilibrium manifolds.Subsequent simulations show that the turbofan engine MRAC method based on the EME model has the same properties as the turbofan engine MRAC method based on the linear model,but the control performance quality is more in line with the engine requirements.At the same time,the MRAC method of high-frequency gain decomposition of the turbofan engine based on the EME model can achieve progressive tracking of the reference model under a ground-wide range of variable operating conditions,but the dynamic performance is poor;when the flight conditions change,the output parameters of engine can remain unchanged under the action of the algorithm,but they cannot reach the specified value under the action of the algorithm.Thirdly,the MRAC method of high-frequency gain decomposition of turbofan engine based on the EME model is not good in the wide range variable condition dynamic performance on the ground,and the MRAC method of high-frequency gain decomposition of turbofan engine with reset mechanism is considered.In the MRAC method of high-frequency gain decomposition of turbofan engine based on the EME model,a reset mechanism is introduced and the output parameters of the reset mechanism are set to adjustable parameters.In the MRAC method of high-frequency gain decomposition of turbofan engine based on the EME model,a reset mechanism is introduced and the output parameters of the reset mechanism are set to adjustable parameters.The theory proves the stability and transient characteristics of the algorithm added to the reset mechanism.The simulation shows that choosing appropriate adaptive gain and adding reset mechanism can improve the system dynamic performance.Finally,in view of the problem that the MRAC method of turbofan engine highfrequency gain decomposition based on the EME model cannot obtain the specified output parameters when the flight conditions change,a switched MRAC method for turbofan engine high-frequency gain decomposition under flight conditions is proposed.The subsystems are divided and different reference models are selected for different subsystems.The stability of the system is analyzed using the multi-Lyapunov method,the transient characteristics of each subsystem are explored,and the absolute value of the error of each subsystem is found to have a maximum value.Using flight condition changes as switching signals,simulation studies have found that the switching MRAC method of turbofan engine high-frequency gain decomposition has a good control effect,which can enable the engine to achieve progressive tracking under different flight conditions to obtain the expected value.Adjustment speed is fast,and dynamic performance is good.Each subsystem has a maximum absolute value of error.