Discussion Forum

Hi everyone:

im modelling interior PMSM in comsol 5.2a for my work at this moment. Firstly i chose soft iron as the material for my stator and rotor and the study solver went on with no problem (both stationary study and transient study). Then when i checked the BH curve of soft iron and realized it isnt as close as the BH stats published by the factory. The 'real' BH curve is much more steeper than the BH curve of soft iron especially in the beginning part. Once i modified the BH curve my study just cannot converge. Two solution i found and gave a shot: a) i used step by step method to ramp up the source (current+PM), but it failed no matter how soft the source is being ramped up. This method is given in details in comsol forum as well. b) i used silicon steel to replace the soft iron since the BH curve for silicon steel is similar to the real BH curve. Then seems like the problem was solved but one thing i noticed is that the transient study goes really slowly.

SO i was wondering how you guys deal with this material nonlinearity and any tips to speed up the transient study? Thank you!

Best Wishes Xu Zhi