I note that SOME solar panel regulators work by shorting the panel (completely, not just loading it) when the desired charge is reached.
That's just ducky for solar panels. The current stays essentially the same, limited by the amount of sunlight striking the panel, while the voltage goes down to the saturation voltage of the transistor (about .4 volts). So the controller stays cool (dissipating about 2 watts for every five amps of panel - possibly less than it dissipates when the current is going through the diode instead) and the panel gets a little more solar heating than it does while it's charging the batteries.
But if you use this on a mill, the genny voltage is controlled by the rotation rate, and that blade is a big flywheel so the speed doesn't change suddenly. When the transistor turns on the current goes up enormously, limited only by the voltage drop in the transistor and the rectifier diodes and the resistance of the coils and wiring. The hundreds, or even thousands, of amps instantly destroy the transistor, and if it fails shorted rather than open (or it takes too long to fail) the rectifier diodes go, too. Those probably fail shorted, and eventually the blade slows to a crawl, after the rest of the momentum is dumped as heat, mostly in the genny windings.
= = = = =
Other types will be OK, or relatively OK: A regulator that opens the circuit won't load the genny (which may overspeed) but won't fry itself and the rest of the system either. A regulator that applies a dump load will do just what you want. (Controllers for large systems should be one of the latter two types. Shunt regulators tend to be for small systems only.)