Parameter Estimation Based Fault Diagnosis in Dynamic Systems

Parameter estimation is a natural approach to the detection and isolation of parametric (multiplicative) faults. In this approach, a reference model is obtained by first identifying the plant in a fault free situation. Then the parameters are repeatedly re-identified on-line. Deviations from the reference model serves as a basis for detection and isolation of different faults. Parameter estimation may be more reliable than the analytical redundancy methods, but it is also more demanding in terms of on-line computation and input excitation requirements.