KDI-Based Robust Fault Detection Scheme for Nearly Linear Systems

抄録

This paper proposes a robust fault detection scheme for nearly linear systems. The system is described by a linear model and the undermodeling error is quantified using a group of fuzzy models. The fault detection is performed by using Kullback discrimination information (KDI) as an index for model discrimination. Our efforts are made to increase its robustness to the undermodeling error.