New approach to model validation and fault diagnosis (Q1367797)

One proposes a new approach to model validation for uncertain systems in which the (nonlinear, time-varying and dynamic) uncertainty is described by an integral quadratic constraint. In this context, the fault diagnosis problem is dealt with as an application provided that the computations can be carried out in real time. The results are based on the solution of a Riccati differential equation and the solution of a set of filtering state equations (closely related to results on observer-based fault detection). The method requires input-output data (not necessarily with zero initial conditions) defined only over a finite time interval.