Fault Detection And Reconstruction For Descriptor Systems Using Sliding Mode Observers

Fault detection and isolation (FDI) is becoming an ever increasingly important research field. A fault is deemed to occur when the system is subject to an abnormal condition, such as malfunctions in actuators or sensors. The fundamental purpose of an FDI scheme is to generate an alarm when a fault occurs and also to identify the nature and location of the fault. Many FDI methods are based on the design of observers. The measured plant output is compared with the output of an observer designed for a model of the controlled system, and the discrepancy is used to form a residual signal. Using this residual signal, a decision is made about whether a fault condition is presented. However, in practical applications, the model of the system possesses uncertainties, which could cause the FDI scheme to trigger a false alarm even when no faults present. On the other hand, the effect of a fault could not be hidden due to the uncertainties, which may go undetected. Hence, there is a real need for robust FDI schemes which are insensitive to model uncertainties.The concept of sliding mode emerged from the Soviet Union in the late sixties where the effects of introducing discontinuous control action into dynamical system were explored. By the use of a judicious switched control law, it was found that the system states could be forced to reach and subsequently remain on a pre-defined surface in the state space. Whilst constrained to this surface, the resulting reduced-order motion, referred to as the sliding motion, was shown to be insensitive to any uncertainties or external disturbance signals. As a result, sliding mode technique has attracted much interest. Particularly, sliding mode technique was subsequently employed in state estimation via an observer.In the paper, sliding mode observer technique is extended to a class of complex systems, that is, generalized dynamic systems. Sliding mode observers are developed to detect and reconstruct simultaneous actuator and sensor faults. The approached approach is settled for minimizing the effect of the uncertainty on the reconstruction of actuator and sensor faults. An example is given to demonstrate the efficiency of the proposed fault diagnostic method.