Limit Management Control For Aero-Engines

With the development of aero-engine technology,the performance of aeroengine is improving.However,the rapid response of the engine often leads to the adverse phenomena such as surge,stall,flameout,over-speed and over-temperature,which seriously affects the engine life.To avoid the above-mentioned situations of the engine in the whole flight envelope and operating condition,and improve the safety of the engine,is limit management problem that needs to be studied in aero-engine control.At present,Min-Max scheme is widely used in aero-engine control method.However,this control structure is too complex to realize in engineering practice,and can not guarantee transient limit protection.In order to solve the problems above,the following studies will be developed.The static characteristics of the Min-Max switching strategy are studied.The activation conditions of the initial controller and the steady-state controller are analyzed in the multi-loop control system.The capacity of the Min-Max strategy to achieve steady-state limit protection is verified.In order to reduce the possibility of limit violation in the transition state,an optimal augmented monotonic tracking control method based on the eigen-structure assignment is proposed.Based on the characteristics of the invariant zeros of augmented system,the sufficient and necessary condition is developed to realize the generalized monotonicity.Genetic algorithm is exploited to optimize the system eigen-structure parameters to obtain generalized monotonicity of key variable of aero-engine.For the problem of slow dynamic response raised by generalized monotonicity,an LMI control method based on ellipsoid invariant set is developed to meet dynamic constraint on the control input and the limit output.Since the constraint on the monotonicity of the limit output is relaxed,the response speed of the engine control is improved.An output-based limit management control method based on bumpless transfer for aero-engines is proposed.In order to establish the relationship between controller switching actions and limit violation,the safety margin function and parameter prediction function are designed to realize the mapping from the outputs to the switching signal,and the switching logic is designed to realize the map from the switching signal to the loop activation index.In order to avoid the engine parameter oscillation caused by the controller switching,based on bumpless transfer method,a compensator is designed to generate a virtual tracking command instead of the real command,which makes the control signal of the inactive loop as close to the control signal of the active loop as possible.In comparison with Min-Max structure,this method can deal with transient limit protection more efficiently and has faster dynamic response speed.Based on terminal set constraint,a model predictive control method with no static error is proposed.The optimal performance index at each sampling time is taken as Lyapunov function,the necessary condition for decreasing the optimal performance index is obtained,the system state is governed into the terminal constraint set defined by the ellipsoid invariant set,and the predictive control problem is transformed into a linear state feedback control problem to stabilize the system,so as to ensure the stability of the predictive control system.A multiplex predictive control method with no static error is proposed.The control input is arranged by designing a multiplexing scheduling law,and finally a lower dimension optimized problem is solved in each time step,so as to reduce the computational complexity and improve the real-time performance of the algorithm.Considering the inevitable degradation of engine performance during the life cycle,a robust reference governor method based on linear state feedback control is proposed,where the health parameter is taken as the disturbance term of the system.When the reference governor has been designed,the SQP algorithm is used to modify the main tracking reference in real time so as to realize limit protection.For nominal system and system with disturbance,the maximum output admissible set is developed to obtain the output admissible index to be used in the SQP algorithm respectively.The simulation results show that this method can realize the robust limit management control of the engine in the case of performance degradation.