High Performance Single and Multiple Faults Diagnosis in Voltage Source Inverter Fed Induction Motor Drives

This paper deals with a new approach for single and multiple open-circuit faults diagnoses in voltage source inverter (VSI) fed induction motors. The diagnosis procedure is based on the knowledge of the outputs inverter currents distribution in alpha-beta frame combined with additional diagnosis variables which use their mean values. In a first step, twenty seven patterns built with output inverter currents in alpha-beta axis are used to detect the faulty legs. Six of them are dedicated to the single-fault modes and the others to the multiple open-switch modes. A second step is achieved by normalizing the average line currents. The use of normalized current-average values allows not only the fast identification of the single fault but also makes it possible to distinguish several multiple-fault cases having the same signatures in alpha-beta frame. The simulation results of fault detection and identification method are included to prove that the proposal leads to good results.