Heuristic Estimation of Parameters in High-Frequency Models of Induction Motors for Bearing Currents Simulation

One of the major disadvantages in the introduction of power electronic devices in the control of electrical machines is the increase of Electromagnetic Interferences (EMI’s). To derive the right models and deep in EMI’s studies some precise simplifications and accurate parameters are essential for credible computer simulations. In this paper, it is proposed a straightforward and flexible parameter estimation tool for High-Frequency (HF) models that permits a complete understanding of EMI phenomena. The presented estimation tool uses a heuristic optimization in order to improve the matching between measurements and simulations of Common Mode (CM) and Differential Mode (DM) impedances. The target function of the proposed formulation is set to minimize the error between the measured and simulated impedances of an induction motor, where different type of errors are considered. In this way, the proposed approach uses an algorithm able to use any heuristic search strategy, to handle different kinds of HF models and to use different comparison methods between measurements and simulations. Finally, as an industrial application, the parameter estimation is used in a model able to simulate bearing currents.