DESIGN AND ANALYSIS OF LINEAR AND NONLINEAR FILTERS FOR THE FDI OF AIRCRAFT MODEL SENSORS

Increasing demands on reliability for safety critical systems such as aircraft or spacecraft require robust control and fault diagnosis capabilities as these systems are potentially subjected to unexpected anomalies and faults in actuators, input-output sensors, components, or subsystems. Consequently, fault diagnosis capabilities and requirements for aerospace applications have recently been receiving a great deal of attention in the research community. A fault diagnosis system needs to detect and isolate the presence and location of the faults, on the basis also of the control system architectures. Development of appropriate techniques and solutions for these tasks are known as the fault detection and isolation (FDI) problem. Several procedures for sensor FDI applied to a nonlinear simulated model of a commercial aircraft, in the presence of wind gust disturbances and measurement errors, are presented in this thesis. The main contributions of this work are related to the design and the optimisation of two FDI schemes based on a linear polynomial method (PM) and the nonlinear geometric approach (NLGA). In the NLGA framework, two further FDI techniques are developed; the first one relies on adaptive filters (NLGA–AF), whilst the second one exploits particle filters (NLGA–PF). The suggested design approaches leads to dynamic filters, the so–called residual generators, that achieve both disturbance decoupling and robustness properties with respect to modelling errors and noise. Moreover, the obtained results highlight a good trade-off between solution complexity and achieved performances. The FDI strategies are applied to the aircraft model in flight conditions characterised by tight–coupled longitudinal and lateral dynamics. The robustness and the reliability properties of the residual generators related to the considered FDI techniques are investigated and verified by simulating a general aircraft reference trajectory. Extensive simulations exploiting the Monte–Carlo analysis tool are also used for assessing the overall performance capabilities of the developed FDI schemes in the presence of both measurement and modelling errors. Comparisons with other disturbance–decoupling methods for FDI based on neural networks (NN) and unknown input Kalman filter (UIKF) are finally reported.