State-space parameter estimation for closed-loop fault detection and identification

This paper focuses on the use of parameter estimation techniques for the implementation of real-time Fault Detection and Diagnosis schemes that are suitable for closed-loop system operation. A description of the main principles regarding fault detection and identification using parameter estimation is given, along with some considerations on the closed-loop system identification problem. The Recursive Least Squares method with exponential weighting and constant forgetting factor is used for the design of the closed-loop system diagnosis procedure. An application example from the field of aircraft control illustrates the suitability of the proposed solution for complex applications where a state-space representation of the system dynamics is used.