Giant kites that fish for wind at high altitude could soon deliver 71% more electricity than anyone expected
Picture a rigid-wing kite the size of a small aircraft, tethered to a ground-based winch and carving figure-eight loops hundreds of meters up — its pulling force converted, through that cable, into electricity. This is airborne wind energy, a technology built to reach winds far stronger and more consistent than any conventional turbine tower can access.
For years, engineers have lacked clear answers on how to fly it — which flap settings, which altitude, which flight path — to get maximum power. A new simulation study is starting to provide those answers, and the numbers are larger than expected.
What is a ground-gen airborne wind energy system?
A ground-gen AWES runs on a straightforward mechanical principle. A rigid-wing kite flies crosswind loops high in the atmosphere. As it pulls against its tether, the cable unspools a winch on the ground, spinning a generator to produce electricity. Once the tether reaches its limit, the kite reels back in using less power than it generated, then the cycle repeats.
Combined with flap scheduling, optimizing this altitude floor brought average reel-out power to 12.96 kW, approximately 71% above the original baseline of 7.56 kW.
The hardware modeled in this study is the Kitemill KM1, a real-world rigid-wing platform. Using an actual commercial system as the reference makes the simulation results more credible and directly applicable to engineering decisions.
Altitude is central to the technology’s appeal. Winds grow stronger, steadier, and less turbulent the higher you go — resources conventional turbines simply can’t reach. The problem is that the relationship between aerodynamic settings, flight-path geometry, and operating altitude hadn’t been well understood under realistic atmospheric conditions. Closing that knowledge gap was precisely the study’s purpose.
Building a simulator rigorous enough to trust
The researchers used a validated nonlinear kite–tether–winch simulator — validated meaning its outputs have been checked against real-world behavior, not just theoretical models. A simulator that hasn’t been tested against physical data can produce convincing but misleading numbers.
To isolate each variable’s individual contribution, the team ran a one-factor-at-a-time campaign, changing one operating parameter while holding everything else constant. Parameters examined included fixed flap deflection, lateral path geometry, wind-field formulation, minimum production height, and the elevation angle of the helix axis.
The wind field was modeled as altitude-dependent, with a time-varying ground-reference wind input — reflecting how real atmospheric winds actually behave rather than assuming a clean, steady flow.
Flap scheduling: A 47% power boost from a simple adjustment
Before any tuning, the baseline system produces an average reel-out power of 7.56 kW. Speed-scheduled flap deflection means adjusting the wing’s flap angle dynamically based on how fast the kite is flying at any given moment. The range tested was conservative: 0° to 10°.
That modest adjustment produced a notable result. Average reel-out power climbed to 11.10 kW — roughly 47% above baseline — while the kite maintained stable crosswind operation throughout. The gain came without disrupting the flight pattern that makes the system work.
A 47% improvement requiring no new hardware is a significant finding, suggesting that many existing systems may be leaving substantial energy on the table.
Flying higher: Another 71% above baseline
Raising the minimum production height — the lowest altitude the kite is permitted to fly during its power-generating phase — pushes it into consistently stronger winds. The aerodynamics of the wing don’t change. The atmosphere does the work.
Combined with flap scheduling, optimizing this altitude floor brought average reel-out power to 12.96 kW, approximately 71% above the original baseline of 7.56 kW. The researchers also tested whether tuning the elevation angle of the helix axis could produce further gains. Within the operating envelope studied, it didn’t improve power output beyond what minimum-height optimization had already achieved.
Path shape matters less than expected
Not every variable proved equally influential. The comparison between circular and elliptical flight paths showed only a minor effect on average traction power — path shape is a lower-priority lever than altitude or flap strategy.
Circular loop radius does have a clear effect on power output, but the overall shape of the loop is largely secondary. For system designers, this narrows the design space considerably. Altitude and flap scheduling deserve attention first; path geometry can be refined once those are locked in.
What comes next for airborne wind energy
These results give the AWES field a clearer map of which operating choices matter most and by how much. The 71% improvement over baseline isn’t a guarantee for every system, but it signals that simulation-guided tuning is a credible path toward significantly better performance.
The immediate next step is testing whether these findings hold up in real flight conditions. The Kitemill KM1 provides a direct bridge between the model and the hardware. If field validation studies reproduce these numbers outside the simulator, the case for airborne wind energy gets considerably stronger.
Reference: Paul, P., Khurshid, Y., Oland, E., Kamra, M. M., and Dief, T. N.: Simulation-Based Assessment of Flap Scheduling and Operating Height for Ground‑Gen Airborne Wind Energy Systems, Wind Energ. Sci. Discuss. [preprint], https://doi.org/10.5194/wes-2026-148, in review, 2026.
Daniel García is an Editor-in-Chief with strong expertise in structural work and engineering principles. He combines this technical foundation with deep knowledge of energy, spatial design, and emerging technologies, bringing a forward-thinking and analytical approach to editorial leadership.
