Dynamic soaring a pumped carousel

AWES whose output drives a carriage around a horizontal ground plane circular rail have a good claim to scaling advantage, as all the heavy equipment is kept on the ground.

The problems associated are the uneven speed and force on the upwind and downwind sides, and the large scale of the ground infrastructure combined with complex controls.

Massimo had a grand course set for kite control to maintain flight and for driving the carriages. Also carriages can be joined around the track to even speed.

Does anyone remember (it’s too early in the morning to research, I’ll edit later) did the Kitegen Carousel plan include any kind of line pumping enhancement mechanism… e.g. Either of
If a carriage was headed upwind on the track the line was let out for height, reeled in on the downwind.
Or vise versa?

L. Fagiano’s thesis is available on https://re.public.polimi.it/handle/11311/1006424.
Click on the green button (Visualliza/Apri) at the bottom right.

Pages 33-35:

II) Variable line length. If line rolling/unrolling is suitably managed during the cycle,
energy can be generated also when the rail vehicle is moving against the wind. In
this case the operating phases of each KSU placed on the HE–carousel, namely
the traction and the unroll phases, are depicted in Figure 2.8. The unroll phase
approximately begins when the angular position Θ of the rail vehicle is such that
the KSU is moving in the opposite direction with respect to the nominal wind:
such situation is identified by angle Θ0 in Figure 2.8. During the unroll phase,
the electric drives linked to the rail vehicle wheels act as motors to drag the KSU
against the wind. At the same time, the kite lines unroll, thus energy is generated
as in the traction phase of the HE–yoyo configuration. The difference between
the energy spent to drag the rail vehicle and the energy generated by unrolling
the lines gives the net energy generated during this phase. When the KSU starts
moving with wind advantage (i.e. its angular position is greater than Θ1 in 2.8),
the HE–carousel traction phase starts: the kite pulls the rail vehicle and the drives
linked to the wheels act as generators. Meanwhile, the kite lines are rolled back
in order to always start the next unroll phase with the same line length. Thus, in
the traction phase the net generated energy is given by the difference between the
energy generated by pulling the rail vehicle and the energy spent to recover the
lines. The MPC controllers employed in the HE–carousel with variable line length
are therefore designed to maximize such a net generated energy.

I don’t know if that replies to the question.