Robust fault detection using inverse images of interval functions

In this paper, a new robust fault detection test based on calculating the inverse image of an interval function is presented. It is based on a consistency test which relies on tools from interval analysis such as set inversion or contractors to check if there exists a member in the family of models described with an interval model that can explain the measured data. The proposed test is compared to related approaches of fault detection using interval analysis. The main features of the new test will be extracted and advantages and drawbacks discussed. An example will be used to illustrate the behaviour.