I've talked with Dr. Green about the parasol idea, we still need to do a stability analysis since L1 is unstable, how much effort will it take to keep the dipole in place given the solar wind can fluctuate by an order of magnitude in either direction of the average. It's a fun problem.

I wonder if having an adjustable center of mass would solve this problem.

A mass ejection method would have to be maintained too often, but CoM adjustments can largely be automated (at least in principle).

The simulation I want to do involves manipulating the dipole strength, since it's deflecting solar wind, there's momentum exchange, if you fluctuate how much area you're blocking out you could get more or less drag as necessary.