A team of researchers has published a new open-access experimental dataset designed to support wind farm control research and numerical model validation. Built from wind tunnel tests using actuated, instrumented, scaled turbine models, the dataset provides time-resolved measurements of turbine loads, actuator commands, and inflow conditions — the kind of controlled, repeatable data the field has largely lacked until now.
What the Dataset Contains
The dataset covers a broad range of turbine-level measurements collected simultaneously during wind tunnel experiments. Researchers recorded tower-base and rotating-shaft moments, rotor speed, generated torque and power, blade pitch angles, nacelle yaw angle, and controller commands. Having all of these channels in a single, time-resolved record is what makes the resource unusual.
Inflow conditions are documented with equal care — wind speed, direction, air density, and wake-flow measurements are all included, giving users a clear picture of what each turbine was actually experiencing at any given moment.
That combination of actuator commands, turbine response, and structural loads captured together is what enables detailed controller and load analysis. Researchers can trace how a specific control input propagated through the turbine structure, which is difficult to do when measurements are partial or averaged over time.
Why the Dataset Was Created
The motivation is straightforward: existing data is not sufficient. Publicly available datasets that provide time-resolved turbine loads, actuator commands, and inflow characterization under controlled operation remain scarce — and that scarcity limits how rigorously numerical models can be benchmarked.
Without reproducible, high-fidelity experimental data, the wind energy community has had to rely on simulations to validate other simulations. It is a circular problem that quietly undermines confidence in model predictions. This dataset directly addresses that gap.
The researchers also frame the release as a contribution to transparency and reproducibility in wind farm design and control research. Open access means other teams can check, extend, and build on the findings rather than starting from scratch.
Wake-Control Strategies Covered
The experiments span a wide range of active wake-control approaches: yaw-based wake steering, curtailment and derating, Helix control, dynamic yaw actuation, Pulse wake mixing, individual pitch control, and several combinations of these strategies.
That breadth is deliberate. Covering multiple approaches under the same controlled conditions lets researchers compare turbine response and controller behavior without confounding variables introduced by different test environments. Wind tunnel conditions are consistent and repeatable — a researcher testing a new control model can run it against the same baseline used in the original experiments, making comparisons meaningful.
Wake control is no longer a single-method discipline. Teams are increasingly exploring hybrid approaches, and a dataset capturing strategy combinations is better positioned to support that direction.
Numerical Models and Benchmarking Framework
The dataset does not stand alone. Numerical models of the experiments accompany it, forming what the researchers describe as a reproducible experimental-numerical benchmarking framework. That pairing allows users to extend the dataset through simulation, run sensitivity analyses, and validate control-oriented aeroelastic and wake-interaction models against a consistent reference.
Fatigue-relevant loading under wake-control operations can also be assessed within this framework. Active control strategies can reduce wake losses, but they may introduce load cycles that affect turbine lifetime — and time-resolved structural load data tied to specific control inputs makes it possible to study that trade-off directly. The paper was published in Wind Energy Science, an open-access journal by Copernicus Publications, so both the article and the dataset are freely available.
Key Takeaways
This release gives wind energy researchers something that has been in short supply: a controlled, time-resolved experimental record linking actuator commands, turbine response, and structural loads across multiple wake-control strategies. It is designed for model validation and benchmarking, and it comes paired with numerical models to support further analysis.
The strategies covered run from yaw steering and derating to Helix, Pulse, and individual pitch control, making the dataset broadly useful across different research directions. Open access removes the remaining barriers.
For teams working on wind farm control algorithms, aeroelastic modeling, or fatigue assessment, this offers a reproducible foundation that was previously difficult to find in the public domain.
Carlos is an engineer with strong expertise in technical and industrial topics. He previously worked at international companies such as Siemens and speaks Spanish, German, English, and Italian.
