Bir elektro hirolik eyleyici sistemin gözlemci tabanli durum denetlenmesi

In this thesis study, jamming fault in an electrohydraulic actuation system which may be quite critical for aerospace applications are detected using model-based approaches. A hydraulic test setup consisting of a connected motor controlled electro hydrostatic actuation (EHA) system and a servo Proportional valve controlled load simulator system is used for the verification of the proposed fault detection and diagnosis (FDD) algorithm. Based on the system dynamics of the equipment in the test bench, mathematical modelling of both EHA and the load simulator is to be performed for further model-based (FDD) techniques. Several model-based approaches proposed for different kind of failure and fault cases are present in the literature. Among them, Kalman filtering and Observer based solutions are the most well-known and preferred model-based methods used in the residual generation step. However, there is no common procedure for the detection and the identification of ‘a fault’ as different failure conditions may result in distinct changes in the nominal behavior of the system or process. Fault to be focused on this study is the jamming phenomena which could occur in an electro hydraulic actuator. This failure case might be severe and even lead to a catastrophic system safety failure. In the scope of the thesis study a dedicated model based approach is developed for the detection and diagnosis of such faulty cases. State and disturbance observer techniques are applied for fault detection purposes. First, the disturbance load acting on an actuator is estimated by a disturbance observer. This observation is correlated with a faulty case and behaved as a residual. Once the residual, the rate of the disturbance estimation, exceeds a predefined value, the fault detection is triggered. The whole procedure is followed with the fault identification step where the moving average of the position tracking error is analyzed to diagnose jamming cases.Thesis (M.S.) -- Graduate School of Natural and Applied Sciences. Mechanical Engineering