Floating wind turbines were built to stay still, but their constant rocking may be quietly rewiring the power grid of the sea
Image created with artificial intelligenceInside a wind tunnel at Politecnico di Milano, two scale-model turbines stand just three rotor diameters apart. The upstream one bobs and tilts on a robotic platform, mimicking the motion of a floating foundation in deep-sea waves. In the invisible column of disturbed air stretching toward the second turbine, something unexpected is happening.
For decades, engineers have treated that rocking motion as a problem to be minimized — a source of mechanical stress and aerodynamic noise. New measurements suggest the platform’s movement may be reshaping the wake in ways that are simultaneously useful and worrying for the turbines downstream.
Why floating turbines move differently — and why it matters for neighbors
A floating wind turbine doesn’t just spin — it also sways, pitches, and yaws on its platform. That motion occurs across two distinct frequency ranges. Slow, eigenmode responses are driven by wind turbulence and can produce large displacements, while faster oscillations, between roughly 0.05 and 0.2 Hz, are driven by ocean waves. Each range produces different aerodynamic effects on the air flowing through and behind the rotor.
Waves arriving at 30° to the wind didn’t consistently improve downstream power, suggesting that motion irregularity limits real-world benefits.
Unlike fixed offshore turbines, a floating machine generates a wake that pulses and meanders in sync with its platform. These coherent flow structures don’t exist in conventional wind farms. Wake losses are already the dominant efficiency problem in any wind farm, and floating motion adds a layer of complexity that engineers are only beginning to quantify.
Inside the experiment: a robotic platform and two model turbines
The researchers used 1:75-scale models of the DTU 10 MW reference turbine, tested in a large wind tunnel at Politecnico di Milano. The upstream turbine was mounted on a six-degrees-of-freedom robotic platform capable of reproducing any combination of floating-platform motion.
Carefully controlled sinusoidal motions were applied — surge, pitch, yaw, and combined surge-sway at various angles to the wind. The team also ran the platform through realistic motion sequences derived from aero-hydro-servo-elastic simulations of a spar-buoy platform responding to irregular waves at 0° and 30° relative to the wind. The downstream turbine stayed fixed, tested at spacings of 3, 4, and 5 rotor diameters and at lateral offsets of zero, half a rotor radius, and one full rotor diameter. Free-stream turbulence intensity was held at 1.5%, representative of low-turbulence offshore conditions.
Power gains: how platform motion wakes up a downstream rotor
In the baseline case — upstream turbine fixed — the downstream turbine produced only 11–18% of upstream power at spacings of 3–5 rotor diameters. A deep, persistent wake deficit was the culprit, driven by a high thrust coefficient (Ct = 0.9) combined with very low ambient turbulence.
Certain motions broke that stalemate. Large-amplitude, low-frequency yaw and crosswind surge-sway motions enhanced lateral wake meandering, boosting downstream power by up to 26% relative to the fixed case. The largest gains appeared in two scenarios: a 3-rotor-diameter aligned layout with yaw motion, and a 5-rotor-diameter aligned layout with surge-sway directed 45° to the wind. That 45° surge-sway increased the downstream turbine’s output by 5.4 W at model scale — equivalent to roughly 0.82 MW at full scale. Waves arriving at 30° to the wind didn’t consistently improve downstream power, suggesting that motion irregularity limits real-world benefits.
The hidden cost: cyclic loads on the waked turbine
The same platform motion that reshapes the wake also sends periodic flow disturbances downstream. The downstream rotor responds with load oscillations locked to the exact frequency of the upstream platform motion.
When the downstream turbine was fully immersed in a pulsating wake — produced by surge or pitch motion in an aligned layout — dynamic thrust oscillations reached approximately 3% of mean thrust, and torque oscillations reached about 12% of mean torque. Those numbers may sound modest, but they repeat at every wave cycle, accumulating fatigue damage over a turbine’s lifetime.
Under realistic wave-driven motion, load oscillations were broader in frequency but smaller in amplitude than in the sinusoidal cases. Motions at higher wave frequencies — corresponding to a reduced frequency of about 1.2 — produced negligible dynamic loads on the downstream rotor, suggesting a natural threshold above which motion-induced wake structures dissipate before reaching the next machine.
What this means for the design of floating wind farms
The study establishes a clear tradeoff. Platform motion can modestly improve downstream energy capture while simultaneously increasing fatigue loading on the waked turbine — and both effects must be weighed together in farm layout decisions and control strategies.
Current industry tools such as FAST.Farm neglect dynamic wake effects from platform motion entirely. The experimental data from this study provide a benchmark to validate and improve such models, filling a gap that numerical simulations alone have struggled to close.
Results do come with caveats. The experiment used low turbulence, short spacings, and a vertically confined wind tunnel. At full-scale spacings of 6–10 rotor diameters, under realistic atmospheric turbulence, the gains may shrink and coherent motion-induced structures may dissipate sooner.
Future work will need to examine whether active blade-pitch and torque control can moderate the dynamic loads identified here, and whether the tradeoff between energy gain and fatigue cost shifts when turbines have the freedom to respond to their own motion. The rocking platform isn’t just a structural challenge — it’s a variable that farm designers will need to actively manage.
If you want to discover more about these findings, you can check the complete study here: Fontanella, A., Cioni, S., Papi, F., Muggiasca, S., Bianchini, A., and Belloli, M.: Experimental investigation of the effects of floating wind turbine motion on a downstream turbine performance and loads, Wind Energ. Sci., 11, 1821–1851, https://doi.org/10.5194/wes-11-1821-2026, 2026.
Carlos is an engineer with strong expertise in technical and industrial topics. He previously worked at international companies such as Siemens and is multilingual.
