Current Harmonics Minimization of PMSM Based on Iterative Learning Control and Neural Networks: Motor Data

The provided motor data corresponds to an electrical machine with 24 stator slots and 16 poles. As is common in electrical machines, this motor generates unwanted flux and current harmonics. However, the accompanying paper presents an effective solution to suppress these harmonics through the combined use of Iterative Learning Control (ILC) and Neural Networks (NNs).

The ILC method demonstrates proficient compensation for harmonics during operations with constant speed and current reference values. Additionally, Neural Networks are trained with data derived from ILC, proving to be highly effective in suppressing harmonics even during transient operation. The simulation model used in the study is based on flux and torque maps, dependent on dq-currents and the electrical angle. These maps are obtained from Finite Element Method (FEM) simulations of an interior permanent magnet synchronous machine (IPM) and are openly published here, intended to facilitate other researchers in making direct comparisons with their own methodologies.

Simulation results presented in the paper confirm that the integration of ILC and NNs leads to superior elimination of current harmonics during transient operations compared to using ILC alone.
If you use the provided maps and motor data, kindly cite the associated paper for reference: https://doi.org/10.3390/machines11080784, https://www.mdpi.com/2075-1702/11/8/784