Self-organizing pattern of magnetic domains I would like to model an structure like this:

using simple methods like cellular automata or networks.
The picture is signed as "a self-organizing pattern of magnetic domains in cobalt".
I'm not a noob in physics; and  understand enough about magnetic domains.
But can not even imagine how this configuration could arise.
Articles on this topic are not available (or are very expensive).
I will be grateful for any ideas and references about the math
underlying such processes.
 A: The specific figure in the original post seems to originate from design and architecture publications, more specifically books of questionable scientific rigorousness. The original source, though, seems to be a paper from the 1950's (e-print, Fig. 5b), and similar-looking pictures can be found in other publications as well (e.g., this paper [e-print] or this PhD dissertation [mirror]).
As for the modeling, i.e., how these patterns can form, it seems that:


*

*a short range ferromagnetism, and

*a longish range (e.g., $\propto r^{-3}$) at least partially anti-ferromagnetic interaction


are the essential ingredients: bubbles and stripes such as this have been shown to arise in ferromagnetic, nearest-neighbor Ising models with dipolar interactions (paper, arxiv; another paper) as well as in spin 1 models with long-range interactions (paper, arxiv, e-print).
With respect to cellular automata or similar models, magnetic-domain-like patterns certainly can be generated (see, e.g., this paper [arxiv] and this [e-print]), as noted by Stephen Wolfram himself (Fig. 13c, remark on page 239), but I didn't find anyone actually modeling the magnetic domains dynamics with cellular automata. That said, magnetic domains have been shown to display features of complex systems at criticality, such as self-organization, avalanches, and fractal geometry.
