Fault detection and isolation in nonlinear dynamic systems: a fuzzy-neural approach

A novel approach based on soft computing concepts is proposed for fault detection and isolation (FDI) of dynamic systems. The proposed method utilizes the concepts of fuzzy clustering, fuzzy decision making and RBF neural networks to create a suitable structure for design of a FDI. The practical applicability is illustrated on a CNC X-axis drive system. Specifically, FDI of twelve different process faults and three different sensor faults is successfully detected for the CNC system.