Thank you @rschmehl.
This confirms what I thought I saw during the projection for a few seconds:
It is interesting to compare this curve with the curve shown in Figure 15 of the publication linked below.
Figure 15
Experimental results from a representative automated power cycle during flight tests at the SkySails Power site in northern Germany in spring 2021 […]. The average power over the pumping cycle is approximately 92 kW.
If the surface area of the kites and their characteristics are similar for these two tests approximately three years apart, we can then see that the average of 92 kW recorded with a wind of approximately 12 m/s could fall within the new curve. Thus these previous tests would also be validated according to the new criterion.
As for this current curve, I have the impression that it sags a little when the wind speed increases. Perhaps, as with conventional wind turbines, the nominal power is approached or even exceeded. Or it could maybe be due to another reason, perhaps linked to the type of performance of a flexible kite in crosswind flight.