Neuro-fuzzy approach for default Diagnosis: Application to the DAMADICS

The failures auto-sensing becomes increasingly essential in the complex systems exploitation. This article consists in working out a system of defects diagnosis based on an artificial intelligence technique which associates fuzzy logic with neural networks. The method is applied to obtain the DAMADICS (Development and Application of Methods for Actuator Diagnosis in Industrial Control Systems). This technique with its capacities of generalization and memorizing gives effective diagnostic tools. In the first part of this work, we studied the modelling of this industrial actuator, simulation gives an idea of the actuator behaviour in a normal mode functioning. Then, we carried out the diagnosis of defects during abnormal operations by using a neuro-fuzzy classifier.