Posters

Abstract

This presentation demonstrates software-based turbine control optimisation achieving >1% AEP gains, validated through a new open-source dataset from Altahullion wind farm combining 1Hz turbine SCADA and LiDAR measurements. Turbine performance was optimised by using a real-time controller continuously evaluating small parameter changes which provides the basis for a continuous and adaptive optimisation method. The presentation will elaborate on the AEP optimisation methodology, uplift measurement and uncertainty analysis. The AEP uplift measurement uses the open-source data and publicly available wind-up Python tool [1]. The presentation aims to progress the industry’s familiarity with AEP optimisation approaches which leverage the existing turbine control system without the requirement of expensive additional hardware; the RES TuneUp product used in this case study provides an example. The presentation also aims to progress the industry’s ability to accurately measure AEP uplift with uncertainty quantification by making public both the data and analysis associated with this case study. The RES performance optimisation method uses a combination of data-driven techniques, classical models and engineering know-how resulting in safe and effective control system changes which optimise AEP. The presentation will show that it is possible to achieve 1%-3% AEP increase on most existing wind farms through a combination of: * Ensuring region 2 fine pitch angle, tip speed ratio and yaw alignment are optimal for each turbine and its unique inflow conditions * Ensuring turbine control setpoints are situation-aware; the optimum can vary with inflow situation (eg high/low shear,  waked vs un-waked wind direction, etc.) * Ensuring yaw control is making the right trade-off between AEP and yaw system duty cycle * Ensuring grid compliance such as reactive power provision is correct * Enhancing AEP further with wind farm control features such as collective control and wake steering (if desired)  The RES performance optimisation method will be demonstrated through the specific example of data-driven optimisation at Altahullion. Innovative aspects of the RES optimisation approach include: * Ability to extract high frequency data and safely change parameters on multiple turbine types (eg Siemens, Vestas) * Use of adaptive control techniques to continuously optimize turbine control * Robust and repeatable uplift measurement using an open method and tool [1] The open-source Altahullion dataset will be a valuable resource to technical stakeholders in the wind industry. Containing 1Hz measurements of all the original SCADA channels and all data from a 2.5D upwind vertical profiling LiDAR, the dataset is applicable to a wide range of research use cases. The process of creating an open-source wind farm dataset and ensuring it has good findability, accessibility, interoperability, and reusability (FAIR) will also be presented. The presentation will conclude by summarizing the achieved performance increase for the open-source Altahullion case study. The performance increase is verified using ground-mounted LiDAR in this case. References [1] wind-up Python package v0.4.5, https://pypi.org/project/res-wind-up/0.4.5/, November 20, 2025