Research On Stability Control For Aero-Engine With Inlet Distortion

The aero-engine suffers much more serious distortion for that advanced tactical aircraft pursues higher stealth and maneuverability performance, which bring much more serious distortion for aero-engine. Now the serious distortion becomes a critical factor to hinder the engine performance improvement. To solve this problem, a new method called distortion tolerate control gets more and more attention, which takes inlet distortion into account, adjusting the engine operating state to satisfy stability requirement and improve the thrust performance under low distortion conditions. Focusing on the stability control with inlet distortion, dynamic modeling, optimization of stability control scheme, LPV control algorithms and stability control system are studied in this thesis.(1) Nonlinear component level model with distortion impact and polynomial LPV models are studied. By mechanism analysis of the distortion impact, an interpolation method of different characteristics depending on distortion index is proposed, which is then imported into a component level model. Combined with the proposed total pressure correction method, the nonlinear component level model with distortion impact is established. Based on linearized models, polynomial LPV models as forms of state space and transfer function are built by polynomial fitting method, which lay the foundation of control system design.(2) Optimization methods for stability control schedule are studied. A modified variable increment LP algorithm is developed for the optimization of control schedule with two variables. The increment coefficient in the LP algorithm is introduced to vary with performance index, which make the optimization algorithm convergence speed increase at the beginning, and in the late optimization provide more accurate results. The simulation results show that this improved LP algorithm can effectively prevent local optimization. The analysis of this method shows that the optimization process is reasonable, and the results obtained are accurate and reliable. The differential evolution optimization method with border buffer zone(DEBZ) is proposed to obtain stability control schedule with four variables. The border buffer zone is used to hold individuals around the constraint boundary, which make evolution more quickly and efficiently. The proposed optimization method shows faster convergent speed and higher convergent accuracy for the optimization problem of multimodal functions with multiple constraints than traditional differential evolution algorithm. By the proposed algorithm, the accuracy of optimized stability control schedule for aero-engine is improved. The resulting control schedule can satisfy the limits under different distortion, and provide maximum thrust performance.(3) Based on generalized distance between systems, a robust gain scheduling control based on generalized distance between systems for polytope LPV system is proposed. The generalized distance between systems is used to solve convex decomposition coefficients, so that the LPV model characterizes controlled object more accurately to improve control accuracy. The simulation results of compressor speed control at the intermediate state demonstrate that the proposed method has better performance than the zoning H ? controller and the robust gain scheduling controller based on parameter geometry distance. To further enhance the control effect in lager operating envelope, a state reset switched polytope LPV approach is proposed for the intermediate state control of a turbofan engine. Through the state reset switch strategy, the designed sub-controllers only need to be efficient in sub-regions, which improve control accuracy and reduce the computational complexity and conservative. Simulation results with a turbofan engine model show that the controller has better performance than the single LPV controller.(4) For the compressor speed which is changed in a large range and controller realizable, a robust LPV/PI control method is proposed. Combining LPV control and PI control structure, robust stability conditions of closed-loop system is deduced by stability analysis method of transfer function model. The resulting conditions are transformed into sum of squares(SOS) programming problem which is easy to be solved. This method can not only provide robust controller, but also has high reliability and good realizability.Gain scheduling control methods based on SOS programming is studied. Gain scheduling control methods based on common Lyapunov function and parameter-dependent Lyapunov function are proposed for aero-engine respectively. Combining bounded real theorem and SOS programming, a common Lyapunov function and a parameter-dependent Lyapunov function are used to analyze the robust stability conditions of closed-loop LPV system, which are then translated into SOS constraints. Simulation results show good dynamic response performance for compressor speed control and engine pressure ratio control.(5) Distortion tolerance control system and engine stall margin adaptive control system are designed, respectively for the sensor-based control mode and model-based control mode. For the distortion tolerance control system, stability control schedule is imported, so as to ensure stable operation of the aero-engine, at the same time get the greatest possible thrust performance. The purpose of engine stall margin adaptive control system is to make the available surge margin track the demand surge margin depending on distortion, so that the control system has the ability to adjust the engine operating state with distortion. Simulation results demonstrate that proper real-time adjustment of engine operating point could meet the fan surge margin requirement, and this system is able to help aero-engine to achieve their full potential.