Skip to Main Content
Fault detection and diagnosis are very much needed in many industrial applications. One of the most popular scheme is the model-based fault diagnostic. Recently, artificial intelligence techniques have been found to be suitable for fault detection and diagnosis and a variety of techniques have been proposed in this area. However, only few applications and real time implementation of the schemes have been reported. In this paper, we use a fault detection and diagnostic scheme based on the model-based approach using parameter estimation and fuzzy inference and experimented it on a DC motor servo trainer. The model of the plant is obtained using the recursive least squares parameter estimation technique, and fuzzy inference is used for the interpretation of the fault. Several faults have been identified on the system. The faults are then simulated on the motor and experiments are carried out to diagnose the types of faults. Experiments show the effectiveness of the proposed technique for real time applications.