Dynamic Simulatoin Study On Throttling Process Of Expander Cycle Rocket Engine

Based on IRC method(a type of centralized parameter method),for 90 kN LH2/LO2 closed-cycle expander rocket engine,this paper introduces the derivation and the equation of every component of liquid rocket engine in system level.A general modular simulation library of liquid rochet engine has been established based on current AMESim libraries and the development by using AMESet tool,such as turbine,centrifugal pump,combuster chamber,etc.Then,a simulation model of 90 kN closed-cycle expander engine has been established,and the model is used for the start-up process of rocket engine.Compared the simulation results with the real engine experiment,the simulation model has been validated.Based on the principle of rocket engine throttling and the system scheme of engine,three throttling methods have been put forward and compared.Then the scheme of no-decoupled double-PID control two turbine bypass valve is selected.In this paper,the transfer function of each part of the control system is derived,and the control system is added into the simulation model.Based on critical proportion method,this paper finds thst the tuned PID parameters is too large.Then a method of double-PID controller parameters tuning has been established,and has been used in 2I2 O engine throttling control system.And this paper analysis the effect of PI parameters value during thrust control process and mass ratio control process.By using the tuned PI parameters,simulations of step/ramp/sine signal have been finished.The result shows that step signal is easy to lead to drastic fluctuation of mass ratio.By using ramp signal,the range of mass ratio fluctuation and the overshoot of thrust reduces with slower ramp signal.The throttling signal speed should be lower than 50% per second to guarantee the mass ratio changement is not exceed +0.10.A method of adaptive Fuzzy-PI control is introduced to improve the engine thorottling dynamic process.This paper uses Simulink to design fuzzy controller,and uses MATLAB-AMESim co-simulation method to establish simulation model.It is found that the dynamic response of Fuzzy-PI controller is better than PI controller.With the optimization of membership functions and fuzzy rules,we can get better dynamic response.