Many reviewers discourage the use of 3D FEA to validate 2D FEA solutions in electrical machine analyses, citing that they both result in similar outcomes. More shockingly, when I try to explain that the 2D FEA models, having end-winding approximations, is performed in static FEA in order to facilitate speed during the design process (mostly with intense optimisation), they insist that experimental measurements, especially where it is difficult to undertake due to many reasons, is the norm. However, I have come to observe that they may be wrong with such a stance given that certain machines, with very high cross-magnetisation, produce fields along their axial paths, which do not correspond their 2-D (static) and 3-D (transient) FEA solutions. Hence, the objective of this inquiry.
I am not expert in electrical machine related work but i am handling finite element related course. What i would like to convey to you that whether the response is 2d or 3d is purely depends on the stress conditions on the machine due to imposed load. If the imposed load on the structure, it does not affect the intermediate stress conditions then it is 2-d problem (Intermediate stress is very minimal), mean only other 2 stress condition is influencing on the response of machine. Same things if intermediate stress influence very much is there in addition to other stresses then it is 3-d problem. So, i hope you understand where do we use 3-d results for comparison in 2-d or 2-d results for comparison in 3-d. I suggest you to read any standard finite element model related text books.
I would also like to tell 1 point that if suppose we wrongly use data for comparison purposes then people are giving critical comment in wrong doing.
3d method takes a lot of time to solve the problem. However it allows to obtain accurate results. 3d and 2d results are not the same. Please refer to my paper 3d and 2d analysis of end effects in linear motors. To validate your Fem results I suggest to use others methods like reluctance network model.
Advances in finite element modelling techniques are closing the gap between 2D and 3D computations for 3D effects. For example, JMag now has a feature that enables one to make computations in 2D, yielding same results as computations in 3D. Of course this is applicable if the machine itself is not inherently a 3D geometry.
- Badges
- Science topic