Electrical tractions use more and more the asynchronous machines because of their

robustness, their power weight ratio. Their maintenance as well as their diagnosis then became an economic state. It is important to early detect the faults likely to appear in those motors and therefor to implement a preventive maintenance.

Thus, a detailed model of the machine was developed in order to analyse the impact of

those faults on the machine performance. We could then get specific signatures for electrical faults and foresee their evolution.

Our approach is based on keeping a close watch over parameters of the performance model of the machine particularly sensitive to faults: The rotor resistances of three phase model, the

magnetising inductance and the rotoric resistance of a two phase model.