Separation of slip- and high-frequency flux densities and its application in rotor iron loss fine analysis of induction motors

Abstract

The fine analysis of rotor losses of induction motors requires separation of the slip- and high-frequency electromagnetic quantities under load conditions. However, the conventional method involves a full slip-cycle finite element simulation, which is very expensive in terms of CPU time and memory usage. To obtain a precise iron loss prediction with reduced time consumption, this paper proposes three methods for fast separation of the slip- and high-frequency electromagnetic quantities at rotor side of an induction motor. The first method (Method A) eliminates the influence of the slip-frequency component by calculating the derivative of the rotor flux density that is obtained by Time-Stepping Finite Element Method (T-S FEM). The second method (Method B) extracts the magnitude of slip-frequency flux density from the spatial distribution of the flux density wave within one supply cycle, and the major high-frequency components are extracted from the wave after subtraction of the full slip-frequency wave determined by Least Square Fitting (LST) the curve in time domain. The third method (Method C) reproduces the full slip-cycle waves by incorporating the wave segments at different positions in rotor teeth, and then separates the slip- and high-frequency components by Discrete Fourier Transform (DFT). The three methods are implemented to separate the slip- and high-frequency rotor flux densities in a 5.5kW induction motor. The accuracy of the results and the CPU time of the three methods are compared with those by the conventional method. The predicted rotor iron losses are also validated by experiments to verify the effectiveness of the presented methods.