High Viability Study Of The Flight Control System

With the development of the aviation technique, higher and higher requirements are presented for the reliability of plane. To meet the needs of them, the Self-repairing Flight Control System (SRFCS) has been developed. The Self-repairing Flight Control System is a new kind of failure tolerant control on the basis of Fly-by-Wire flight control system. The Fault detection and diagnosis (FDD) of the flight control system is one of the important parts of the SRFCS. The active failure tolerant control and the failure diagnosis constitute the major part of the analytical redundancy.This thesis, based on one UAV's pre-research, tackles on the application of the fault detection and diagnosis in the flight control system. Failure detection filter, Parity Space and Strong Tracking Filter theory which is a nonlinear fault diagnosis method are applied to the fault detection and diagnosis of the control surface in this thesis. The residual decision combined with wavelet analysis is discussed. The sensor and ruder fault detection and isolation are also studied. The main content in the thesis is as follows.1. The nonlinear dynamic model of the plane is built for the thesis and the fault models of the control surface, executor (rudder) and the sensor (gyro) are established as well.2. The application of the parity space, fault detection filter in the control surface fault detection and diagnosis is studied. A new arithmetic for the parity space theory to get the fault parameter is presented. Based on the theory analysis, fault diagnosis is simulated and the same result with academic analysis is gotten.3. The application of Strong Tracking Filter, which is a nonlinear fault diagnosis method in the fault diagnosis of the control surface is discussed. The formula and arithmetic of the fault diagnosis are derived. Simulation results illustrate this arithmetic can detect the control surface fault quickly and get the fault parameter with high accurate.4. The above three fault diagnosis methods are analysed, and then it is pointed that the parity space arithmetic is more useable and reliable to the control surface fault diagnosis of the High Altitude and Long Flight Time UAV.5. The fault detection and isolation of the primary sensor, which is gyro, is discussed in the thesis. Methods are put forward which are used for unorthodoxy placement gyro fault detection and isolation when the three gyros have trouble by the attitude signal.6. The rudder's fault detection of the plane is studied. The Improved Sequential Probability Ratios is used to improve the detection speed and the detect result can be got in ten steps. The System Identification theory is adopted to diagnose fault.7. The wavelet analysis is applied to deal with the residual signal, since the residual signal includes the non-stationary process. The simulation result is satisfied.The fault diagnosis methods of the plane are studied and improved in the thesis. The simulation results show the effectiveness of these methods. It is believed that it will be useable to the High Altitude and Long Flight Time UAV.