Set-membership methodology for model-based systems prognosis

This paper addresses unknown input interval estimation and prognosis for a class of uncertain systems. Under the assumption that the measurement noise and the disturbances are bounded, lower and upper bounds for the unmeasured state and unknown inputs are computed. Then, damage state estimation is formulated as a set-inversion problem. The set-membership methodology is applied to an electromechanical oscillator to show the effectiveness of the proposed technique.