How to run a skew optimization study?


This article shows how to use MANATEE software dedicated to the electromagnetic and vibro-acoustic simulation of electric motors to carry a fast multiobjective optimization of the skew pattern. Skewing can indeed help reducing both acoustic noise due to magnetic forces and torque ripple, but the objectives might be contradictory and a multi-objective optimization requires significant computing time.


Skewing is an important technique to reduce electromagnetically-excited noise and vibrations in electrical powertrains. However, it reduces the average torque, it may increase radial force harmonic at some specific operating points, and more importantly it may not minimize acoustic noise and torque ripple at the same time. It is therefore necessary to quickly estimate the effect of skewing on these quantities. This can be carried in MANATEE software using a sensitivity study or a special optimization project but a more efficient technique consists in carrying skew optimization study after the calculation of flux lookup tables. This "optimization" consists in simply sweeping all skewing values specified by the user as calculation is very fast. The study results in the following graphical output:

Example of skew optimization study carried on a PMSM in MANATEE
Example of skew optimization study carried on a PMSM in MANATEE

Scripting implementation

To automatically run the skew study at the end of the calculation the simulation project should include Input.Simu.is_skew_optim = 1;

The maximum rotor skew angle (in stator slot pitch) is specified with Input.Simu.skew_rater_max = 2;

The total number of skewing values to be investigated is given by Input.Simu.Nb_skew = 20;

The skew optimization environment can also be launched after a basic simulation of MANATEE by calling the following command line: >>run_skewoptim

GUI implementation


See also

Skew optimization is illustrated in validation projects ALG_034 and ALG_035.

Previous Next