Core power losses characterization based on inverter/induction machine design

Induction machines and the adjustable speed applications are a well established reality of todays industry. In a growing domain, adjustable speed is being utilized not only for industrial processes but for energy generation as well. However the performance of the induction machine is different either if it is fed by a static power converter or if it is fed directly from the grid. Surely, there are not technological limitations, but imposed limitations of a standard construction, for generic use at constant voltage and frequency. Traditional design methodologies are not able to satisfy these new application demands what implies to establish a new integrated methodology based on integrating simultaneous analysis of the whole system. This way the constraints imposed on the machine by standard design can be overcome. The proposed methodology in this paper combines the design of the power converter operating within real electrical conditions with the finite element design of the machine, in order to optimize machine and overall operation. It gives a contribution to the knowledge of converter and machine interactions in the whole system and carries out a better characterization of the power losses what reflects in a more efficient thermal sizing and reduces the overrating of the machine.