A note on nonparametric kernel smoothing for model-free fault symptom generation

This note describes a novel approach to the generation of fault symptoms in nonlinear systems. The method is model-free, in that no analytical model o f a plant is needed. The kernel smoother makes it possible to detect abrupt changes in the plant dynamics, possibly due to some malfunction. A simple sufficient condition for fault detectability is proved and experimental res ults on a real benchmark are reported. (C) 1999 Elsevier Science Ltd. All r ights reserved.