I am working on a coreless axial flux permanent magnet machine design, and am having issues with the transformation of magnetic equivalent circuits using 2D models.
A 3D magnetic system can be modeled as a 2D magnetic system only approximately. In this case, it is carried out in two stages, where at each stage the magnetic system is modeled for 2 dimensions, and the third dimension is taken as homogeneous (monotonic).
For example, let's define the measurements as follows: X - radial direction, Y - tangential direction, Z - axial direction. The tangential direction turns into a line.
At the first stage, we model the section in the XZ plane. The Y direction is assumed to be uniform. The goal is to determine the width of the magnetic system in the X direction if the non-uniform field is replaced by its uniform counterpart.
At the second stage, we model the section in the XY plane. The X direction is considered uniform.
To increase accuracy, the process can be repeated several times.