Wind turbines have been quietly stealing power from each other for decades and engineers just found a smarter way to make them step aside
Every wind turbine leaves a scar in the air behind it — a cone of slow, churning wind that can quietly rob the next turbine in line of a meaningful share of its output. Multiply that across hundreds of machines in a modern wind farm, and the losses add up fast. Engineers have long tried to outsmart this by tilting turbines slightly off the wind, nudging their wakes sideways before they can reach downstream rotors. But predicting exactly what happens next has proven stubbornly difficult. A new analytical model may have finally changed that.
The hidden cost of turbines working too close together
When a wind turbine spins, it doesn’t just generate electricity — it also leaves a trail of depleted, turbulent air stretching far downwind. Any turbine sitting in that trail receives weaker, choppier wind, which directly cuts into its power output. Across a large wind farm, these wake interactions compound into one of the largest sources of energy loss operators face.
The engineering fix is called yaw-based wake steering. By deliberately rotating a turbine slightly off the wind direction — a condition called yaw misalignment — engineers can nudge its wake sideways, away from downstream rotors. The concept is well established and has been tested in real farms. Modeling it accurately across an entire array, where wakes overlap, merge, and influence each other in complex ways, is the harder problem.
Because the model is analytical and non-iterative, it slots naturally into the low-order control and optimization frameworks that wind farm operators already use.
Why existing models stumble on a chain reaction called secondary wake steering
Most engineering models handle multi-turbine farms by stacking individual turbine wake models on top of each other. That works reasonably well for simple cases but breaks down when interactions grow more complex.
Secondary wake steering is one particularly tricky phenomenon. An upstream turbine tilts off the wind, redirecting its wake sideways. A downstream turbine isn’t tilted at all — yet its own wake still gets deflected, because it sits inside the cross-flow generated by the upstream machine. Capturing this accurately has required either computationally expensive iterative calculations or rough approximations, and neither approach suits the fast, low-order models that real wind farm control systems depend on.
A new model built on momentum — without the computational cost
The core innovation extends a “conservation of momentum deficit” approach to the lateral direction. Previous momentum-based models focused on the streamwise direction — along the wind. This new framework adds a lateral component, producing a fully analytical solution that requires no iterative loops.
Separating the streamwise and lateral parts of the wake makes the physics of yaw-induced deflection traceable. Wake deflection is derived by integrating the local skew angle at each point, grounded directly in fluid mechanics. The practical payoff is speed: no iterations means the model can run fast enough for real-time wind farm control and optimization, fitting naturally into the low-order frameworks that operators actually use.
What the model revealed about secondary wake steering
Beyond computational efficiency, the model offers a clearer physical picture of what happens inside a steered wind farm. For an unyawed downstream turbine, its wake is deflected by two distinct forces acting together.
The first is the cross-flow arriving from the upstream tilted turbine. The second is subtler — the downstream turbine’s own streamwise velocity deficit slows the air carrying that cross-flow, amplifying the deflection effect. According to the model, roughly one-third of the downstream wake’s deflection comes from this self-deficit contribution — a mechanism that previous models didn’t explicitly account for. Making that mechanism visible gives engineers not just a better number, but a better explanation.
Tested against wind-tunnel data — and outperforming established models
The researchers validated the model against wind-tunnel measurements covering both single-turbine and multi-turbine configurations. It accurately reproduced far-wake deflection and lateral velocity distributions across the tested cases, outperforming established engineering models in most configurations.
The standout result came when comparing performance across the full hub-height velocity field. Unlike metrics that rely on detecting a wake’s center point — which becomes ill-defined when wakes merge — this field-based comparison stays well-defined in all conditions. On that measure, the new model was more accurate in every configuration tested. Its main limitation is the near-wake region, where the simplified treatment underestimates the streamwise velocity deficit, a known trade-off the researchers acknowledge directly.
What this means for the future of wind farm control
Because the model is analytical and non-iterative, it slots naturally into the low-order control and optimization frameworks that wind farm operators already use. A model that’s physically accurate but too slow to run in real time has limited practical value. This one is designed to be both.
Better wake-steering models could meaningfully increase the energy yield of existing wind farms without adding a single turbine. As farms grow larger and turbines are packed more densely to make the most of available land and offshore sites, wake-interaction losses grow with them — making accurate, fast models increasingly valuable. The model’s transparent structure also makes it easier to identify where to extend it, with the near-wake limitation a clear target for future work.
Reference: Gaukroger, N. J., Siguenza-Alvarado, D., van der Laan, M. P., and Bastankhah, M.: An analytical wake deflection model for wind farms including secondary wake steering, Wind Energ. Sci. Discuss. [preprint], https://doi.org/10.5194/wes-2026-183, 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.
