Robust analytical redundancy relations and artificial neural networks for fault detection and isolation in electric drives

Abstract

In this paper, a fault diagnostic system in electric drives using a robust analytical redundancy relations and perceptron multilayer artificial neural network based on classifier method. The ARRs and ANN design process are clearly described. For this purpose, we have treated the signals of the measured parameters (current and speed) to use them firstly. The simulation model of the electric drives is studied under normal and different fault mode (short-circuit and open faults of switching device in inverter) in electric drives are considered for fault detection and diagnosis. Robustness of classifier load torque is verified. Analysis, modelling and simulation results are presented to demonstrate the validity of the proposed method.