Presentations

Abstract

Wind farm flow control is a promising solution to mitigate the wake losses within a wind farm, with wake steering being the most mature technique. Recently, active wake mixing control strategies have gained increasing attention within the research community due to their high potential in terms of power gains. Instead of redirecting the wake away from the downstream turbines, these strategies enhance the mixing with the surrounding free-stream flow to accelerate wake recovery. An example of a wake mixing strategy is the helix approach, in which a sinusoidal signal is imposed on the pitch angle of the blades. However, the commercial deployment of wake mixing strategies is challenged by their associated increase in the structural loads.  In this work, we evaluate the potential of three different control strategies in a large offshore wind farm. The strategies considered here are wake steering, the helix method, and a novel combined strategy in which a subset of turbines operates under wake steering while others apply the helix control. A wind direction uncertainty of 2.5 degrees is considered to provide a realistic estimate of the annual energy production (AEP) gains achievable by these strategies. The control setpoints of each strategy are optimized for each turbine and flow case. The control optimization is formulated as a multi-objective problem, whose objective function includes both power production and a weighted penalty on control effort. This latter term can serve as proxy for structural loads, especially for the helix operation. By varying the weight on the control effort penalty, both aggressive and conservative strategies can be explored, obtaining a Pareto front of optimal solutions. This method enables a systematic investigation of the trade-offs between increased power production and required control effort by the control strategies. The methodology is applied to a scaled version of the Hollandse Kust Noord (HKN) offshore wind farm in the Netherlands, comprising of 69 IEA 22 MW turbines. The results show that the novel combined wake steering-helix strategy outperforms the individual strategies for the large wind farm considered as case study. This performance improvement arises from the complementary effectiveness of the two techniques in enhancing power production. Overall, an AEP gain of up to  0.88% is obtained for the combined strategy under these realistic uncertainty assumptions. However, this AEP gain is associated to control operation times up to 60% for some turbines, raising concerns regarding the required control effort, and consequently on the structural loads.  Therefore, a trade-off solution is identified using the multi-objective approach, yielding an AEP gain of  0.80% while limiting the average control operation time to 22%.  A more detailed analysis of the control setpoints reveals the influence of incorporating a penalty on control effort. As the penalty weight increases, the wind farm control strategy increasingly concentrates on turbines located at the boundaries of the farm. Conversely, control actions applied to turbines in the central region are substantially reduced, indicating that the majority of the power increase is contributed by boundary turbines.