Research On Robust PID Controller For Aero-engine

The purpose of aero-engine control system design is to achieve the satisfactory steady state anddynamic performances for all of the flight conditions. Aero-engine is a kind of nonlinear systems anddynamic characteristics are various at different operation points and flight conditions. PID controlleris widely used for aero-engine. In this paper, the PID parameters tuning method is studied to yield arobust performance for model uncertainties in full flight envelope.A model which includes aero-engine, hydraulic servo actuators and controller was constructedusing the co-simulation of AMESim and Simulink. The linear models of rotator speed system,pressure ratio system, fuel actuator and nozzle actuator were established. These models havesatisfactory accuracy in comparison with the nonlinear models.The robust PID controller based on internal model control (IMC-PID) was presented. The PIDparameters can be fixed to a filter coefficient λ with PID principle approaching IMC. The filtercoefficient which is a parameter of the feedback filter can meet the demands of robustness inuncertain conditions. The maximum magnitude of the closed-loop frequency response (Mp) was usedas a dynamic performance criterion. The method of Mp tuning criterion to select a set of tuningparameters λ was presented. With this tuning criterion, the controllers could be designed to meetdesired performance and robustness as close as possible over all uncertain processes. An improvedmethod was presented to adjust the tuning parameters of PID controller when model error is large.With IMC-PID and theory of similarity, PID parameters in full flight envelope were designed.The PID control problem was expressed in a linear-quadratic-regulator framework with statefeedback which can be solved using linear matrix inequality (LMI). The poles of the closed loopsystem are placed in the cycle region of complex plan for system robustness. The relationship of thetransient response of the system and pole placement is analyzed. The specific circle regionsrepresented as a restriction of LQR. LMI descriptions from different specifications were used todesign a robust PID controller. Simulations results show that the robust PID controller was adaptableto the uncertain system parameters.PID parameters of pressure ratio control system were tuned using IMC-PID and genetic algorithm,respectively. The controller designed by IMC-PID has better dynamic performances in comparisonwith that the controller designed by genetic algorithm.