Conference Paper

Simultaneous sensor and actuator fault reconstruction and diagnosis using generalized sliding mode observers

Dept. of Electr. & Comput. Eng., Univ. of Alberta, Edmonton, AB, Canada
Conference: American Control Conference (ACC), 2010
Source: IEEE Xplore

ABSTRACT A new filter for state and fault estimation in a class of nonlinear systems is presented in this paper. The observer benefits from both sliding mode control and singular systems theory. The novelty of this approach is based upon dealing with systems prone to faults at sensors and actuators during the course of the system's operation coincidentally. Conditions and proofs of conversion for the proposed observer are presented. A noticeable feature of the proposed approach is that the state trajectories do not leave the sliding manifold even in presence of sensor/actuator faults. This allows for actuator faults to be reconstructed based upon information retrieved from the equivalent output error injection signal. Due to employing a generalized state space form (singular system theory), the sensor faults are also estimated.

1 Follower
  • [Show abstract] [Hide abstract]
    ABSTRACT: This paper proposes a method for robust and simultaneous actuator and sensor faults reconstruction of linear uncertain system based on sliding mode observer (SMO). In comparison with existing work, the observer contains two discontinuous terms to solve the problem of simultaneous faults. The idea is to introduce an appropriate filter on the systems output to transform the sensor faults in the fictitious actuator faults and hence the proposed SMO can be applied.
    Electrical Engineering and Software Applications (ICEESA), Hammamet; 03/2013
  • [Show abstract] [Hide abstract]
    ABSTRACT: Detection, isolation, and reconstruction of multiple system, actuator, and sensor faults in time-delayed linear systems are considered in this note. Based on an appropriate change of system coordinates; all system, actuator, and sensor faults are assigned to different subsystems of a time-delayed linear system. Then, using sliding mode observers and equivalent output error injection approaches, all faults are reconstructed separately. This results in reduced computational complexity. Finally, numerical simulation is used to show the feasibility and effectiveness of the proposed approach.
    Control and Automation (ICCA), 2011 9th IEEE International Conference on; 01/2011
  • [Show abstract] [Hide abstract]
    ABSTRACT: A multiple fault tolerant measurement system based on nonlinear dynamic models, special searching algorithm, principle components decomposition and Q test is developed. The proposed system uses a model-based estimator to deliver symptoms. The symptoms are then analyzed in a statistical unit in order to detect the faults and isolate the faulty sensors. Multi-layer perceptron networks, radial basis function networks and Tagaki–Sugeno fuzzy models were examined for the fault estimator module and among these fuzzy models presented the best performance. The main advantages of the proposed scheme are the capability to detect, isolate and repair multiple faults in both input and output sensors and the feasibility to be applied to any system with as many sensors as required, all due to particular design of its model-based estimator. The system was tested on a CSTH model developed based on an experimental platform; different experiments demonstrated satisfactory results.
    Neural Computing and Applications 12/2013; 23(S1). DOI:10.1007/s00521-012-1328-0 · 1.76 Impact Factor


Available from