Automated Parameter Identification for Multiple Coupled Circuit Modeling of Induction Machines

The parameterization of a multiple coupled circuit model for induction machines requires either precise knowledge of the internal structure and materials in an electrical machine or complex measurement and test procedures. In the presented approach, the parameters for the modeling are automatically determined and optimized with measurement data from standard operation using a differential evolution algorithm. This method is primarily oriented towards practical suitability for industrial applications. The multilevel approach enables the emulation of the dynamic behavior of unknown induction motors. The parameterized modeling can then be used for the detection and diagnosis of fault cases.