Glad to hear the simulation converged.
You don't want to make a 3D simulation because you will lose the symmetry boundary condition that you got for free in the axisymmetric model, and that missing BC might give problems with convergence in 3D.
You don't need to make a 3D simulation. Simply export two directional deformation results, one for X and one for Y, as text files. Those will come with the nodal X, Y coordinates as well as the deformation for each node in X and Y. Convert that 2D radial set of points into a 3D cloud of points by rotating the points in 15 degree increments about the Y axis. You should be able to import that into your optical simulation software.
Please reply with the name of the Optical Simulation software you are using.
If you can insert an image of the lens assembly cross-section, that would be appreciated.