Application of wake steering to offshore wind farms

Project Description:

Prior research has shown that turbine wakes are largely responsible for power losses in wind farms [R1]. In this project, we aim to improve the performance of offshore wind farms by implementing a new Wind Farm Control (WFC) strategy. Existing wind farms operate based on the separate control of turbines with the goal of maximising the energy generation of individual turbines. So-called greedy turbines within the first rows of an array extract a high amount of energy from incoming wind with considerably less energy available for those downstream. This leads to significant power losses in wind farms [R2].

The primary need of wind farm developers/operators is reliable WFC strategies that can effectively mitigate wake effects. WFC strategies are based on the notion of controlling the whole wind farm as a single unit.

Based on this approach, a wind farm is perceived as a network in which turbines can communicate with each other [R3]. Among different WFC strategies, wake steering is believed to be the most promising technique: some turbines are intentionally operated in yawed conditions to steer their wakes away from downwind turbines. Although this reduces the power generated by yawed turbines, the power generated by the wind farm as a whole is increased [R4]. Wake steering is, however, a new field of research, that has not been employed widely within the wind energy industry. The aim of this project is to develop a fast- running wake steering model that can predict and optimise the efficiency of offshore wind farms in real time. The PhD student will collaborate with the US’s National Renewable Energy Laboratory (NREL) in the use of high-fidelity flow simulations to calibrate the model to be developed, for various offshore conditions and farm layout geometries.

The outcome of this project will lead to a promising WFC strategy to reduce the levelised cost of energy (LCOE) for offshore wind turbine arrays, including those with floating platforms. In doing so will narrow the gap between research and practice, while contributing towards addressing important wind-farm optimisation challenges that the wind energy industry currently faces.

Methodology

To predict power production in wind farms, a common research practise is to use advanced high-fidelity numerical and experimental techniques to characterise wake effects. However, in real-life applications, so many different scenarios need to be tested due to the continuous change of operating conditions (e.g., change in wind direction angle and wind speed magnitude) [R5]. This means that the wind energy industry cannot often afford the methods used in the research community.

Fast-running models are therefore required to develop an effective wake steering strategy suitable for the wind energy industry. To develop such models, a deep understanding of complex physics of steered wakes is crucial. In this project, we will exploit fundamental fluids mechanics and turbulence research to develop a simple, yet reliable, physics-based wake steering model. Due to its simplicity and ease of use, it will enable wind farm designers/operators to reliably deploy it in real time without the need of a deep knowledge of the fluid mechanics or numerical computing underpinning it. High-fidelity wind flow simulation data provided by the NREL collaborators will be used to calibrate the developed model. The simulation data will be interrogated to study characteristics that impact the performance of wind farms and wake propagation, and how they vary between offshore locations. The calibrated wake model will be used to optimise the performance of offshore wind farms with different layout geometries for several UK offshore locations.