Muti-mode Switching Control For Aircraft Engines

With the development of aircraft engine control systems, intelligent engine technology hasbecome an important direction. Through the "health" management, intelligent engine real-timemonitors the changes of performance and environmenting, switches to the corresponding controlmode. Therefore, it would greatly improve engine performance, reliability and preparation, andextend the life of engine. Multi-mode switching control technology, as one of key technologies ofintelligent engine, gets increasing attention and become a research hotspot. Focusing on the subject ofmulti-mode switching control of aircraft engine, dynamic modeling, multi-mode switching algoritm,fault diagnosis and fault-tolerant control are studied in the thesis. The main work and contributionsare as follows:（1）Nonlinear component level model and state variables linear model are established of aturbofan engine. A mixture solution method based on variable scale method is proposed to establishstate variables linear model. This proposed method avoids the disadvantages such as less annuracy ofsmall perturbation method, inconsistent of dynamic process of steady-state value of response, and lessaccuracy and longer time of traditional fitting method with the increase of the matrix dimensions. Thelinear model is in good agreement in the dynamic response, and steady accuracy with componentlevel model.（2）For the jump of parameters and the unstable problem of the system during the shift betweendifferent control modes, a fuzzy switching control scheme is proposed for aircraft engine controlsystem. Fuzzy switching controllers for different structures of control modes are designed. Thestability of the multi-mode switching control system for aircraft engines is analyzed based on thepassivity theorem. Sufficient stable condition of multi-mode switching system is given. Simulationresults show that the proposed method can not only ensure the system stable but also have favorableperformance in the flight envelop.（3）The problem of multi-section optimization and switch control in acceleration process ofaircraft engine are researched. The acceleration process is divided into three phases. The objectiveindexes and constraints are different in each phase. The optimal acceleration control law is obtainedusing the SQP algorithm to optimize each acceleration phase. For the significant jump and theexceeding of parameter limits during the shift of different phases on the acceleration process, a fuzzyswitching controller based on the cloud model is derived. Simulation results show that the proposedmethod can not only enhance acceleration performance but also guarantee the smooth transition of acceleration response.（4）The knowledge-based and model-based methods are used respectively to design aircraftengine components fault and sensors fault diagnosis and isolation system. The knowledge-based cloudrelational analysis is proposed which integrates the cloud theory and the relational analysis. A faultdiagnosis system utilizing the cloud relational analysis is designed based on the analysis to the faultcharacteristics of the sensors and the engine components. The proposed method can not only isolatesensor fault and engine component fault, but also locate the fault position of the sensors or thecomponents. Kalman filter based fault diagnoses system designed for aircraft engine component andsensor faults. A double-threshold mechanism is proposed to select the threshold in senor faultdiagnosis. This greatly improves the accuracy and robustness of sensor fault diagnosis system. Theaircraft engine component and sensor fault detection and isolation system is established. The approachcan not only distinguish among sensor fault, component fault but also isolate and locate sensor faultand engine component fault when the two kinds of faults occur simultaneously. The fault diagnosisaccurency is effectively improved for the component-sensor simultaneous fault.（5）Integrated engine control and fault diagnoses method, engine fault-tolerant control schemebased control mode switch is designed, including task-level mode and engine-level mode. Aimed atengine components fault, the performance recovery control system of task-level mode had beendesigned. The system switched among the conventional speed control mode, steady state performancerecovery control mode and acceleration performance recovery control modes, based on the engineoperation condition and healthy condition. Aimed at engine control project in the idle state, throttlestate and intermediate state, when a control loop failure occurs, the control strategies switch to othercontrol loop according to fault conditions, to ensure the engine control system performance upgradingsensor fault-tolerant control from hardware and software redundancy to the integrated engine controland fault diagnoses level.