The research work done in this PhD has been carried out under an industrial convention (CIFRE) between the IMS laboratory (Bordeaux University, France) and Airbus Operations S.A.S. (Toulouse, France). The thesis deals with two important Electrical Flight Control System failure cases: runaway (a.k.a. hard over) and jamming (or lock-in-place failure) of aircraft control surfaces. Early and robust detection of such failures is an important issue for achieving sustainability goals and for early system reconfiguration. The thesis focuses on the elevator runaway and jamming. Three model-based monitoring strategies are presented. The first approach is based on a dedicated Kalman filtering with optimised tuning parameters. The second method is based on a decision test applied to an identified sensitive direction in the parametric space. Finally, the third solution is based on a sliding mode differentiator. The techniques have been implemented in the flight control computer and validated on Airbus test facilities and during real flight tests. The experimental results confirmed that good level of performance and robustness can be obtained.
