I don't think I've seen a graph like yours with the W-hour info on
top.

boB

I suspect that the wind power duty cycle is normally very low and the alternators are working nearer a duty cycle associated with small welders and other intermittently rated devices.

With your heating load you have the option to match the load far better to the alternator in higher winds so stator temperature will be much less of an issue.

Also as there is negligible heat in low winds it doesn't matter much if you are well off the best operating point in low winds. A high start up speed and running in stall at the lower wind speeds will cost you very little in terms of heat produced.

If you can keep control of the thing by furling in high winds then you may not need much in the way of ideal load matching. A single fixed load is possibly being very optimistic but you may manage with it.

The thing to avoid at all cost is the temptation to add excess load in high wind to keep speed down rather than furl. If you have to do that then do it drastically enough to stall the prop and reduce power into the alternator. Don't work at the point where the prop is not stalled and you are dissipating more in the windings than into the load.

I see you are looking at capacitor series circuits. Yes, fine in theory but at low voltage and low frequency I wouldn't consider it a viable solution unless you have a source of large quantities of near free capacitors. The cost in capacitors to do it satisfactorily will be ridiculous. Electrolytics I would consider a short term solution. the only saving thing is that electrolytics generally fail to short and would be safe for the machine. They also sometimes fail spectacularly to mechanically open circuit with a bang and lots of mess so keep them in a closed container if you do play with them.