You can find a simple explanation of what an induction generator can do. It is totally self-synchronizing with the grid it is feeding power into. It stays in sync and delivers energy as long as it is driven at an rpm higher than its synchronous speed. If it is feeding a 50 Hz power supply it must be driven 2% to 5% above ts sync speed. Sync speed is 120 X frequency / number of poles thus a 4 pole machine on 50 Hz power supply must be driven at 1500 plus 30 rpm min and 75 rpm ideal. Speed range 1530 to 1575.
The output of the induction generator will stay locked exactly to the grid it is driving into over that speed range. In other words gents the sync is independent of speed.
Now if you drive it with an induction motor connected to another grid which has its own variations in frequency. The induction motor's speed will follow the frequency variations of the grid it is connected to and the induction motors speed variations will pass mechanically to the induction generator BUT THOSE SPEED VARIATIONS WILL HAVE NO EFFECT ON THE SYNCHRONIZATION OF THE INDUCTION GENERATOR with the grid it is feeding power to. About over speed do not apply here because it is being driven by an induction motor that cannot over speed. For those of you who want to see the math there is plenty covering the subject on the internet just ask Google.
The grid feeding powering the induction motor can be a 60 Hz power supply or in a HVDC system a DC motor as long as the gearing connecting it to the induction generator is such that the induction generator is running in the correct speed range it will deliver power in sync. In the case of a four pole induction generator feeding a 50 Hz power supply it will deliver max power when its speed is 1575 to 1600 rpm. The driving motor must have sufficient torque to drive it at that speed against the back EMF coming from the grid it is feeding.
As to the questions of power capability it is better to use many such machine combinations coupled in parallel to one bus-bar. By matching motor and generator efficiency of > 75 % is easily achieved while both machines are operating close to full load rating.
I can tell all of you that the harmonics caused by semiconductor converters are causing serious problems in all sorts of places and it will only be a few years before every one has to have a rethink about using them. Grids that are being fed by large number of wind generators are already finding that the problems are becoming intolerable. It has become fashion to have wind generators feed the grid via semiconductor conversion equipment, in some instances the harmonics are being generated in the RF range and 50 Hz and 60 Hz systems can be badly affected.