Multivariable Robust Controller Design For Aeroengines

With the improvement of the performance of modern airplane, more and more maneuverability requirements of aeroengine have been offered. However, aeroengines are very complicated nonlinear multivariable time-variant aerothermodynamic systems. If there is no guarantee of control system, the aeroengine can not meet required performance requirements, even can not work normally. In order to guarantee the working stably and safety of an aeroengine, meeting the optimum performances under any conditions of flying, it is necessary to investigate the control system of aeroengine deeply and constantly. Currently the controller of aero-engine is mainly two variables, with the continuous improvement of the engine performance requirements, two-variable controllor is clearly not sufficient to allow the engine to obtain a satisfactory performance, so it is necessary to increase variables, in order that the control system can meet the higher requirements. Nearly two decades, robust control is a focal point because it not only can deal with model uncertainty and have mature algorithms, but also controller has robustness. It has made lots of achievement on theory and application.In this paper, X turbofan engine is controlled device, to build the four-variable model of the aeroengine,∞robust control method is used to design control system of the aeroengine. And dynamic response at different condition is achieved. So the main work of this thesis includes:1. Analysis the component-base model, learn the working process of every part and determin the flight envelope. Given the fact that there are too many variables of aeroengine, and different variables have varying performances. Under the premise of considering the power loading and heat loading, on the condition of full using of tunable variables and easy measurement, A four-variables small perturbation state variable model of aeroengine is estabilished according to the improved least square method in the flight envelope, fitting curves show the builded model possesses enough accuracy. The analysings of the model show the uncertainty and the necessity of controller.2. Considering the uncertainty of model and the disturbance in flying, the mixed sensity H∞design method is used to the aeroengine. The H∞controller is obtained by careful selection of weighting matrixes. The simulation results show that, the designed control system not only meets required performance requirements at the design point, but also possesses robust stability and robust performance within the entire flight envelope, even poccesses the ability of anti-interfernce.3. The performances of controller depend on the weighting matrixes, against the dependence on the experience of designers, the design method based on genetic algorithms is used to the aeroengine. Calculate aim function using design aim, the controller is obtained by optimization of weighting matrixes, which solves the problem that normal methed is laborious and time-consuming. The simulation results show that, the designed control system not only meets required performance requirements at the design point, but also possesses robust stability and robust performance within the entire flight envelope, even poccesses the ability of anti-interfernce.4. The simulation results show that the feasibility of the method and the designed aeroengine control system satisfies the required performance requirements.