In this paper it is shown how the well-known linearization technique can be used to solve the problem of fault diagnosis, for non-linear systems subject to actuator failures. An exponential adaptive observer is designed on the basis of the first approximation approach, achieving locally exponential convergence of observation error as well as unbiased fault diagnosis. A link between linear fault diagnosis and its non-linear counterpart is explored. The effectiveness of the proposed non-linear fault diagnosis scheme is illustrated through a simulation example.
BassevilleM.BenvenisteA.Detection of Abrupt Changes in Signals and Dynamic Systems, Lecture Notes in Computer and Information Science, 1985 (Springer-Verlag, Berlin).
2.
ChowE. Y.WillskyA. S.Analytical redundancy and the design of robust failure detection systems. IEEE Trans. Autom. Control, 1984, 29, 603–614.
3.
FrankP. M.Enhancement of robustness in observer-based fault detection. Int. J. Control, 1994, 59, 955–981.
4.
IsermannR.Process fault diagnosis based on modelling and estimation methods—a survey. Automatica, 1984, 20, 387–404.
5.
ParkJ.RizzoniG.An eigenstructure assignment algorithm for the design of fault detection filters. IEEE Trans. Autom. Control, 1994, 39, 1521–1524.
6.
PattonR. J.FrankP.ClarkR.Fault Diagnosis in Dynamic Systems: Theory and Application, Prentice-Hall International Series in Systems and Control Engineering, 1989 (Prentice-Hall, Englewood Cliffs, New Jersey).
TrunovA.PolycarpouM. M.Robust fault diagnosis of state and sensor faults in nonlinear multivariable systems. Proceedings of the American Control Conference, Philadelphia, Pennsylvania, 1998, pp. 608–612.
