FAULT DIAGNOSIS ON INDUSTRIAL SYSTEMS BASED ON A MULTIPLE MODEL APPROACH
Résumé
In this paper, fault diagnosis of industrial systems based on an adaptive filter is developed. Based on the assumption that dynamic behavior of the process is defined as a multiple model approach established around different operating points, a set of residual vector insensitive to faults is established and thus generating a global model exploited in an adaptive filter. The proposed method makes possible fault detection, isolation and estimation for the whole operating range of industrial system. Convergence and stability conditions of the adaptive filter are developed and its performances are tested in simulation.