Table of contents

The 52 papers in this volume constitute the proceedings of the 2017 Wake Conference, held in Visby on the island of Gotland, Sweden. The Wake Conference series began in Visby, where it was held in 2009 and 2011. In 2013 the conference took place in Copenhagen where it was combined with the International Conference on Offshore Wind Energy and Ocean Energy. In 2015 it went back to where it started, Visby, and this time it once again takes place at Uppsala University's Gotland campus, May 30^th - June 1^st.

Modern wind turbines are today clustered in large farms with a total production capacity reaching those of a nuclear power plant. When placed in a wind farm, the turbines will be fully or partially influenced by the wake of upstream turbines. This wake interaction results in a decreased power production, caused by the lower kinetic energy in the wind, and an increase in the turbulence intensity. Therefore, understanding the physical nature of vortices and their dynamics in the wake of a turbine is important for the optimal design of wind farms.

The increased importance and interest in the field of wake and wind farm aerodynamics can be seen in the increased number of scientific articles on the subject. For example, on the Web of Science citation index, the number citations on the topic 'wind turbine wakes' increased from about 50 in 2006 to more than 3800 in 2016. This citation growth essentially shows that the growth in the global production of electrical energy has become a scientific problem to be solved by scientists and engineers. In order to make a substantial impact on one of the most significant challenges of our time, global climate change, the wind industry's growth must continue. A part of making this growth possible will require research into the physics of wind turbine wakes and wind farms.

This conference is aimed at scientists and PhD students working in the field of wake dynamics. The conference covers the following subject areas: Wake and vortex dynamics, instabilities in trailing vortices and wakes, simulation and measurements of wakes, analytical approaches for modeling wakes, wake interaction and other wind farm investigations.

We would like to thank all those that assisted in the review process for these proceedings and we are also thankful for the help we have recieved from the editorial team at IOP Publishing during the process.

Visby, Sweden May/June 2017

Andrew Barney, Stefan Ivanell, Maria Klemm and Jens Nørkær Sørensen

All papers published in this volume of Journal of Physics: Conference Series have been peer reviewed through processes administered by the proceedings Editors. Reviews were conducted by expert referees to the professional and scientific standards expected of a proceedings journal published by IOP Publishing.

Ten simulations of large wind farms have been performed using a fully coupled LES and aero-elastic framework to form a database of full turbine operational conditions in terms of both production and loads. The performance is examined in terms of averaged power production and thrust, as well as 10min equivalent flapwise bending, yaw, and tilt moment loads. Certain scenarios operating below rated wind speed shows unexpected peaks in the loads. The influence on the operating conditions are examined for various parameters and compared relative to an effective power production per area.

We present a wind simulation framework for offshore and onshore wind farms. The simulation framework involves an automatic hybrid high-quality mesh generation process, a pre-processing to impose initial and boundary conditions, and a solver for the Reynolds Averaged Navier-Stokes (RANS) equations with two different turbulence models, a modified standard k- model and a realizable k- model in which we included the Coriolis effects. Wind turbines are modeled as actuator discs. The wind farm simulation framework has been implemented in Alya, an in-house High Performance Computing (HPC) multi-physics finite element parallel solver. An application example is shown for an onshore wind farm composed of 165 turbines.

We present Large-Eddy Simulation results of a turbine wake in realistic complex terrain with slopes above 0.5. By comparing simulations including and without the wind turbine we can estimate the induction factor, a, and we show how the presence of a strong recirculation zone in the terrain dictates the positioning of the wake. This last finding is in contrast to what would happen in gentle terrain with no substantial increase of turbulent kinetic energy in the terrain induced wakes.

With the increase of installed wind power capacity, the contribution of wind power curtailment to power balancing becomes more relevant. Determining the available power during curtailment at the wind farm level is not trivial, as curtailment changes the wake effects in a wind farm. Current best practice to estimate the available power is to sum the available power calculated by every wind turbine. However, during curtailment the changed local wind conditions at the wind turbines lead to inaccurate results at the wind farm level. This paper presents an algorithm to determine the available power of a wind farm during curtailment. Moreover, results of curtailment experiments are discussed that were performed on nearshore wind farm Westermeerwind to validate the algorithm. For the case where a single turbine is being curtailed, it is shown that the algorithm reduces the estimation error for the first downstream turbine significantly. Further development of the algorithm is required for accurate estimation of the second turbine. All further downstream turbines did not experience a change in wake conditions.

Experimental data acquired in the New MEXICO experiment on a yawed 4.5m diameter rotor model turbine are used here to validate the actuator line (AL) and actuator disc (AD) models implemented in the Large Eddy Simulation code EllipSys3D in terms of loading and velocity field. Even without modelling the geometry of the hub and nacelle, the AL and AD models produce similar results that are generally in good agreement with the experimental data under the various configurations considered. As expected, the AL model does better at capturing the induction effects from the individual blade tip vortices, while the AD model can reproduce the averaged features of the flow. The importance of using high quality airfoil data (including 3D corrections) as well as a fine grid resolution is highlighted by the results obtained. Overall, it is found that both models can satisfactorily predict the 3D velocity field and blade loading of the New MEXICO rotor under yawed inflow.

This work deals with the experimental analysis of wake losses fluctuations at full-scale wind turbines. The test case is a wind farm sited on a moderately complex terrain: 4 turbines are installed, having 2 MW of rated power each. The sources of information are the time-resolved data, as collected from the OPC server, and the 10-minutes averaged SCADA data. The objective is to compare the statistical distributions of wake losses for far and middle wakes, as can be observed through the "fast" lens of time-resolved data, for certain selected test-case time series, and through the "slow" lens of SCADA data, on a much longer time basis that allow to set the standards of the mean wake losses along the wind farm. Further, time-resolved data are used for an insight into the spectral properties of wake fluctuations, highlighting the role of the wind turbine as low-pass filter. Summarizing, the wind rose, the layout of the site and the structure of the data sets at disposal allow to study middle and far wake behavior, with a "slow" and "fast" perspective.

This paper is aimed to obtain a better understanding of how a stably stratified atmospheric condition involving low-level jet, in comparison to neutral condition, affects to a wind-turbine performance. In this work, large-eddy simulation approach in combination with actuator-line model is used to simulate the wake flows in an infinitely long wind farm. Here, numerical simulations of the wind-turbine wakes are carried out under the influence of neutral and stable conditions. The numerical results are compared between the two atmospheric conditions. It has been seen that the wind-turbine performance is highly influenced under the stable conditions. For an infinitely long wind-farm scenario, the power produced by a turbine under the stable condition is smaller by 81% than that under the neutral condition.

Increasing use of wind energy over the years results in more and larger clustered wind farms. It is therefore fundamental to have an in-depth knowledge of wind-turbine wakes, and especially a better understanding of the well-known but less understood wake-meandering phenomenon which causes the wake to move as a whole in both horizontal and vertical directions as it is convected downstream. This oscillatory motion of the wake is crucial for loading on downstream turbines because it increases fatigue loads and in particular yaw loads. In order to address this phenomenon, experimental investigations were carried out in an atmospheric-boundary-layer wind tunnel using a 3 × 3 scaled wind farm composed of three-bladed rotating wind-turbine models subject to a neutral atmospheric boundary layer (ABL) corresponding to a slightly rough terrain, i.e. to offshore conditions. Particle Image Velocimetry (PIV) measurements were performed in a horizontal plane, at hub height, in the wake of the three wind turbines in the wind-farm centreline. From the PIV velocity fields obtained, the wake-centrelines were determined and a spectral analysis was performed to obtain the characteristics of the wake-meandering phenomenon. In addition, Hot-Wire Anemometry (HWA) measurements were performed in the wakes of the same wind turbines to validate the PIV results. The spectral analysis performed with the spatial and temporal signals obtained from PIV and HWA measurements respectively, led to Strouhal numbers St = fD/U_hub ≃ 0.20 − 0.22.

3D Reynolds-averaged Navier-Stokes (RANS) simulations of a fully developed wind farm boundary layer over a staggered array of NREL 5MW turbines are presented. The turbine is modeled as an actuator disk and as a fully resolved rotor to compare the effect of the turbine model on the wind farm aerodynamics, in particular the streamwise momentum balance across the farm. Results show that the difference in the turbine model affects the average wind speed through the farm as well as the local flow pattern around each turbine; both contributing to the difference in the prediction of farm performance. Results are also compared with a simple theoretical model of very large wind farms proposed recently. The actuator disk simulations agree very well with the theoretical model, whereas the fully resolved rotor simulations show some consistent and expected differences from the model.

The main objective of this work is to estimate how much of the discrepancy between measured and modeled flow parameters can be attributed to wake effects. The real case simulations were performed for a period of 15 days with the Weather Research and Forecasting (WRF) model and nested down to a Large-Eddy Simulation (LES) scale of ∼ 100 m. Beyond the coastal escarpment, the site is flat and homogeneous and the study focuses on a meteorological mast and a northern turbine subjected to the wake of a southern turbine. The observational data set collected during the Prince Edward Island Wind Energy Experiment (PEIWEE) includes a sonic anemometer at 60 m mounted onto the mast, and measurements from the two turbines. Wake versus free stream conditions are distinguished based on measured wind direction while assuming constant expansion for the wake of the southern turbine. During the period considered the mast and northern turbine were under the southern turbine wake ∼ 16% and ∼ 11% of the time, respectively. Under these conditions, the model overestimates the wind speed and underestimates the turbulence intensity at the mast but not at the northern turbine, where the effect of wakes on the model error is unclear and other model limitations are likely more important. The wind direction difference between the southern and northern turbines is slightly underestimated by the model regardless of whether free stream or wake conditions are observed, indicating that it may be due to factors unrelated to the wake development such as surface forcings. Finally, coupling an inexpensive wake model to the high-fidelity simulation as a post-processing tool drives the simulated wind speeds at the mast significantly closer to the observed values, but the opposite is true at the coastal turbine which is in the far wake. This indicates that the application of a post-processing wake correction should be performed with caution and may increase the wind speed errors when other important sources of uncertainty in the model and data are not considered.

The aerodynamics of Vertical Axis Wind Turbines (VAWTs) is inherently unsteady, which leads to vorticity shedding mechanisms due to both the lift distribution along the blade and its time evolution. In this paper, we perform large-scale, fine-resolution Large Eddy Simulations of the flow past Vertical Axis Wind Turbines by means of a state-of-the-art Vortex Particle-Mesh (VPM) method combined with immersed lifting lines. Inflow turbulence with a prescribed turbulence intensity (TI) is injected at the inlet of the simulation either from a precomputed synthetic turbulence field obtained using the Mann algorithm [1] or generated on the-fly using time-correlated synthetic velocity planes. The wake of a standard, medium-solidity, H-shaped machine is simulated for several TI levels. The complex wake development is captured in details and over long distances: from the blades to the near wake coherent vortices, then through the transitional ones to the fully developed turbulent far wake. Mean flow and turbulence statistics are computed over more than 10 diameters downstream of the machine. The sensitivity of the wake topology and decay to the TI and to the operating conditions is then assessed.

The aim of the present paper is to obtain a better understanding of long distance wakes generated by wind farms as a first step towards a better understanding of farm to farm interaction. The Horns Rev I (HR) wind farm is considered for this purpose, where comparisons are performed between microscale Large Eddy Simulations (LES) using an Actuator Disc model (ACD), mesoscale simulations in the Weather Research and Forecasting Model (WRF) using a wind turbine parameterization, production data as well as wind measurements in the wind farm wake. The LES is manually set up according to the wind conditions obtained from the mesoscale simulation as a first step towards a meso/microscale coupling.

The LES using an ACD are performed in the EllipSys3D code. A forced boundary layer (FBL) approach is used to introduce the desired wind shear and the atmospheric turbulence field from the Mann model. The WRF uses a wind turbine parameterization based on momentum sink. To make comparisons with the LESs and the site data possible an idealized setup of WRF is used in this study.

The case studied here considers a westerly wind direction sector (at hub height) of 270 ± 2.5 degrees and a wind speed of 8 ± 0.5 m/s. For both the simulations and the site data a neutral atmosphere is considered. The simulation results for the relative production as well as the wind speed 2 km and 6 km downstream from the wind farm are compared to site data. Further comparisons between LES and WRF are also performed regarding the wake recovery and expansion.

The results are also compared to an earlier study of HR using LES as well as an earlier comparison of LES and WRF. Overall the results in this study show a better agreement between LES and WRF as well as better agreement between simulations and site data.

The procedure of using the profile from WRF as inlet to LES can be seen as a simplified coupling of the models that could be developed further to combine the methods for cases of farm to farm interaction.

Wind farm control, in which turbine controllers are coordinated to improve farmwide performance, is an active field of research. One form of wind farm control is wake steering, in which a turbine is yawed to the inflow to redirect its wake away from downstream turbines. Wake steering has been studied in depth in simulations as well as in wind tunnels and scaled test facilities. This work performs a field test of wake steering on a full-scale turbine. In the campaign, the yaw controller of the turbine has been set to track different yaw misalignment set points while a nacelle-mounted lidar scans the wake at several ranges downwind. The lidar measurements are combined with turbine data, as well as measurements of the inflow made by a highly instrumented meteorological mast. These measurements are then compared to the predictions of a wind farm control-oriented model of wakes.

Lidar velocity measurements need to be interpreted differently than conventional in-situ readings. A commonly ignored factor is "volume-averaging", which refers to lidars not sampling in a single, distinct point but along its entire beam length. However, especially in regions with large velocity gradients, like the rotor wake, can it be detrimental. Hence, an efficient algorithm mimicking lidar flow sampling is presented, which considers both pulsed and continous-wave lidar weighting functions. The flow-field around a 2.3 MW turbine is simulated using Detached Eddy Simulation in combination with an actuator line to test the algorithm and investigate the potential impact of volume-averaging. Even with very few points discretising the lidar beam is volume-averaging captured accurately. The difference in a lidar compared to a point measurement is greatest at the wake edges and increases from 30% one rotor diameter (D) downstream of the rotor to 60% at 3D.

Nowadays, not only the size of single wind turbines but also the size of wind farms is increasing. Understanding the interaction between the turbines and especially the wakes formed behind them are getting more important to further improve such wind turbine arrays. Consequently, new issues in wind energy research arise. An experimental wind tunnel study was conducted, in order to analyze and understand the far wake of a wind turbine. The experimental results were used to test if an analytical wake model derived by H. Schilichting for blunt bodies can be used to describe the velocity and width development in the far wake of wind turbines. The results of the evaluation show that the wake of a wind turbine agrees fairly well with the model according to Schlichting. The velocity deficit as well as the width in the wake behind the turbine, are found to deviate with around only 2% from the results obtained applying the analytical model. Thus, it can be concluded that the analytical wake model by Schlichting is well suited to estimate the velocity deficit and the width in the far wake of a wind turbine.

The present paper aims at describing the use of a Synthetic-Eddy-Method (SEM), initially proposed by Jarrin et al. [12], in the 3D Lagrangian Vortex method framework. The SEM method is used here in order to generate a far-field incoming flow with a prescribed ambient turbulence intensity. However, for the account of the diffusive term in the Navier-Stokes equations, a classical Particle Strength Exchange model with a LES eddy viscosity is used.

Firstly, the general characteristics of the Synthetic-Eddy-Method will be presented together with its integration in the framework of the developed 3D unsteady Lagrangian Vortex software [27]. The capability of the ambient turbulence model to reproduce a perturbed flow that verifies any turbulence intensity I_∞ and any anisotropic ratio (σ_u:σ_v:σ_w) will be discussed and validated. Then, the capability of the presented ambient turbulence model to compute turbine wakes will also be presented together with first results. Finally, comparisons will be made between the obtained numerical results against experimental data [22, 23] for two levels of ambient turbulence, namely I_∞ = 3% and I_∞ = 15%. Although the present study was initially performed in the framework of tidal energy, its application to wind energy is straightforward.

The wake produced by a wind turbine is dynamically meandering and of rather narrow nature. Only when looking at large time averages, the wake appears to be static and rather broad, and is then well described by simple engineering models like the Jensen wake model (JWM). We generalise the latter deterministic models to a statistical meandering wake model (SMWM), where a random directional deflection is assigned to a narrow wake in such a way that on average it resembles a broad Jensen wake. In a second step, the model is further generalised to wind-farm level, where the deflections of the multiple wakes are treated as independently and identically distributed random variables. When carefully calibrated to the Nysted wind farm, the ensemble average of the statistical model produces the same wind-direction dependence of the power efficiency as obtained from the standard Jensen model. Upon using the JWM to perform a yaw-angle optimisation of wind-farm power output, we find an optimisation gain of 6.7% for the Nysted wind farm when compared to zero yaw angles and averaged over all wind directions. When applying the obtained JWM-based optimised yaw angles to the SMWM, the ensemble-averaged gain is calculated to be 7.5%. This outcome indicates the possible operational robustness of an optimised yaw control for real-life wind farms.

Recently, the Actuator Line Model (ALM) has been evaluated with coarser resolution and larger time steps than what is generally recommended, taking into account an atmospheric sheared and turbulent inflow condition. The aim of the present paper is to continue these studies, assessing the capability of the ALM to represent the wind turbines' interactions in an onshore wind farm. The 'Libertad' wind farm, which consists of four 1.9MW Vestas V100 wind turbines, was simulated considering different wind directions, and the results were compared with the wind farm SCADA data, finding good agreement between them. A sensitivity analysis was performed to evaluate the influence of the spatial resolution, finding acceptable agreement, although some differences were found. It is believed that these differences are due to the characteristics of the different Atmospheric Boundary Layer (ABL) simulations taken as inflow condition (precursor simulations).

Airborne Wind Energy (AWE) is an emerging technology in the field of renewable energy that uses kites to harvest wind energy. However, unlike for conventional wind turbines, the wind environment in AWE systems has not yet been studied in much detail. We propose a simulation framework using Large Eddy Simulation to model the wakes of such kite systems and offer a comparison with turbine-like wakes. In order to model the kite effects on the flow, a lifting line technique is used. We investigate different wake configurations related to the operation modes of wind turbines and airborne systems in drag mode. In the turbine mode, the aerodynamic torque of the blades is directly added to the flow. In the kite drag mode, the aerodynamic torque of the wings is directly balanced by an opposite torque induced by on-board generators; this results in a total torque on the flow that is zero. We present the main differences in wake characteristics, especially flow induction and vorticity fields, for the depicted operation modes both with laminar and turbulent inflows.

Because the horizontal homogeneity assumption is violated in wakes flows, lidars face difficulties when reconstructing wind fields. Further, small-scale turbulence which is prevalent in wake flows causes Doppler spectrum widths to be broader than in the free stream. In this study the Doppler peak variance is used as a detection parameter for wakes. A one month long measurement campaign, where a continuous-wave lidar on a turbine has been exposed to multiple wake situations, is used to test the detection capabilities. The results show that it is possible to identify situation where a downstream turbine is in wake by comparing the peak widths. The used lidar is inexpensive and brings instalments on every turbine within economical reach. Thus, the information gathered by the lidars can be used for improved control at wind farm level.

High-resolution lidar wake measurements are part of an ongoing field campaign being conducted at the Scaled Wind Farm Technology facility by Sandia National Laboratories and the National Renewable Energy Laboratory using a customized scanning lidar from the Technical University of Denmark. One of the primary objectives is to collect experimental data to improve the predictive capability of wind plant computational models to represent the response of the turbine wake to varying inflow conditions and turbine operating states. The present work summarizes the experimental setup and illustrates several wake measurement example cases. The cases focus on demonstrating the impact of the atmospheric conditions on the wake shape and position, and exhibit a sample of the data that has been made public through the Department of Energy Atmosphere to Electrons Data Archive and Portal.

The presence of conservative forces on rotor blades is neglected in the blade element theory and all the numerical methods derived from it (like e.g. the blade element momentum theory and the actuator line technique). This might seem a reasonable simplification of the real flow of rotor blades, since conservative loads, by definition, do not contribute to the power conversion. However, conservative loads originating from the chordwise bound vorticity might affect the tip vortex trajectory, as we discussed in a previous work. In that work we also hypothesized that this effect, in turn, could influence the wake induction and correspondingly the rotor performance.

In the current work we extend a standard actuator line model in order to account for the conservative loads at the blade tip. This allows to isolate the influence of conservative forces from other effects. The comparison of numerical results with and without conservative loads enables to confirm qualitatively their relevance for the near wake and the rotor performance. However, an accurate quantitative assessment of the effect still remains out of reach due to the inherent uncertainty of the numerical model.

In this work, wind-turbine wakes are studied over flat and hilly terrains. Measurements made by using stereoscopic PIV are compared to data obtained from numerical simulations using RANS equations and an actuator-disc method. The numerical and experimental data show similar qualitative trends, indicating that the wind-turbine wake is perturbed by the presence of the hills. Additionally, a faster flow recovery at hub height is seen with the hilly terrain, indicating that the hills presence is beneficial for downstream turbines exposed to wake-interaction effects. The Jensen wake model is implemented over the hilly terrain and it is shown that this model cannot accurately capture the wake modulations induced by the hills. However, by superimposing a wind-turbine wake simulated over flat terrain on the hilly-terrain flow field, it is illustrated that the commonly-used wake-superposition technique can yield reasonable results if the used wake model has sufficient accuracy.

This paper uses a year of SCADA data from Whitelee Wind Farm near Glasgow to investigate wind turbine wake development in moderately complex terrain. Atmospheric stability measurements in terms of Richardson number from a met mast at an adjoining site have been obtained and used to assess the impact of stability on wake development. Considerable filtering of these data has been undertaken to ensure that all turbines are working normally and are well aligned with the wind direction. A group of six wind turbines, more or less in a line, have been selected for analysis, and winds within a 2 degree direction sector about this line are used to ensure, as far as possible, that all the turbines investigated are fully immersed in the wake/s of the upstream turbine/s. Results show how the terrain effects combine with the wake effects, with both being of comparable importance for the site in question. Comparison has been made with results from two commercial CFD codes for neutral stability, and reasonable agreement is demonstrated. Richardson number has been plotted against wind shear and turbulence intensity at a met mast on the wind farm that for the selected wind direction is not in the wake of any turbines. Good correlations are found indicating that the Richardson numbers obtained are reliable. The filtered data used for wake analysis were split according to Richardson number into two groups representing slightly stable to neutral, and unstable conditions. Very little difference in wake development is apparent. A greater difference can be observed when the data are separated simply by turbulence intensity, suggesting that, although turbulence intensity is correlated with stability, of the two it is the parameter that most directly impacts on wake development through mixing of ambient and wake flows.

Previous attempts to describe the structure of wind turbine wakes and their mutual interaction were mostly limited to large-eddy and Reynolds-averaged Navier–Stokes simulations using finite-volume solvers. We employ the higher-order spectral-element code Nek5000 to study the influence of numerical aspects on the prediction of the wind turbine wake structure and the wake interaction between two turbines. The spectral-element method enables an accurate representation of the vortical structures, with lower numerical dissipation than the more commonly used finite-volume codes. The wind-turbine blades are modeled as body forces using the actuator-line method (ACL) in the incompressible Navier–Stokes equations. Both tower and nacelle are represented with appropriate body forces. An inflow boundary condition is used which emulates homogeneous isotropic turbulence of wind-tunnel flows. We validate the implementation with results from experimental campaigns undertaken at the Norwegian University of Science and Technology (NTNU Blind Tests), investigate parametric influences and compare computational aspects with existing numerical simulations. In general the results show good agreement between the experiments and the numerical simulations both for a single-turbine setup as well as a two-turbine setup where the turbines are offset in the spanwise direction. A shift in the wake center caused by the tower wake is detected similar to experiments. The additional velocity deficit caused by the tower agrees well with the experimental data. The wake is captured well by Nek5000 in comparison with experiments both for the single wind turbine and in the two-turbine setup. The blade loading however shows large discrepancies for the high-turbulence, two-turbine case. While the experiments predicted higher thrust for the downstream turbine than for the upstream turbine, the opposite case was observed in Nek5000.

The aim of the present paper is to demonstrate the capability of medium fidelity modelling of wind turbine component fatigue loading, when the wind turbines are subjected to wake affected non-stationary flow fields under non-neutral atmospheric stability conditions. To accomplish this we combine the classical Dynamic Wake Meandering model with a fundamental conjecture stating: Atmospheric boundary layer stability affects primary wake meandering dynamics driven by large turbulent scales, whereas wake expansion in the meandering frame of reference is hardly affected. Inclusion of stability (i.e. buoyancy) in description of both large- and small scale atmospheric boundary layer turbulence is facilitated by a generalization of the classical Mann spectral tensor, which consistently includes buoyancy effects. With non-stationary wind turbine inflow fields modelled as described above, fatigue loads are obtained using the state-of-the art aeroelastic model HAWC2.

The Lillgrund offshore wind farm (WF) constitute an interesting case study for wind farm model validation, because the WT interspacing is small, which in turn means that wake effects are significant. A huge data set, comprising 5 years of blade and tower load recordings, is available for model validation. For a multitude of wake situations this data set displays a considerable scatter, which to a large degree seems to be caused by atmospheric boundary layer stability effects. Notable is also that rotating wind turbine components predominantly experience high fatigue loading for stable stratification with significant shear, whereas high fatigue loading of non-rotating wind turbine components are associated with unstable atmospheric boundary layer stratification.

This paper presents a comparison between measured and simulated tower loads for the Danish offshore wind farm Nysted 2. Previously, only limited full scale experimental data containing tower load measurements have been published, and in many cases the measurements include only a limited range of wind speeds. In general, tower loads in wake conditions are very challenging to predict correctly in simulations. The Nysted project offers an improved insight to this field as six wind turbines located in the Nysted II wind farm have been instrumented to measure tower top and tower bottom moments. All recorded structural data have been organized in a database, which in addition contains relevant wind turbine SCADA data as well as relevant meteorological data – e.g. wind speed and wind direction – from an offshore mast located in the immediate vicinity of the wind farm. The database contains data from a period extending over a time span of more than 3 years. Based on the recorded data basic mechanisms driving the increased loading experienced by wind turbines operating in offshore wind farm conditions have been identified, characterized and modeled. The modeling is based on the Dynamic Wake Meandering (DWM) approach in combination with the state-of-the-art aeroelastic model HAWC2, and has previously as well as in this study shown good agreement with the measurements. The conclusions from the study have several parts. In general the tower bending and yaw loads show a good agreement between measurements and simulations. However, there are situations that are still difficult to match. One is tower loads of single-wake operation near rated ambient wind speed for single wake situations for spacing's around 7-8D. A specific target of the study was to investigate whether the largest tower fatigue loads are associated with a certain downstream distance. This has been identified in both simulations and measurements, though a rather flat optimum is seen in the measurements.

In this paper, an onshore wind farm in mountainous area of southwest China was investigated through numerical and experimental methods. An improved actuator disk method, taking rotor data (i.e. blade geometry information, attack angle, blade pitch angle) into account, was carried out to investigate the flow characteristic of the wind farm, especially the wake developing behind the wind turbines. Comparing to the classic AD method and the situ measurements, the improved AD shows better agreements with the measurements. The turbine power was automatically predicted in CFD by blade element method, which agreed well with the measurement results. The study proved that the steady CFD simulation with improved actuator disk method was able to evaluate wind resource well and give good balance between computing efficiency and accuracy, in contrary to much more expensive computation methods such as actuator-line/actuator-surface transient model, or less accurate methods such as linear velocity reduction wake model.

In Southern Germany a test site will be erected in complex terrain. The purpose is to enable detailed scientific studies of terrain impact on the characteristics of two research wind turbines and to demonstrate new technologies. Within preparatory studies an appropriate site was identified and examined by field tests and numerical studies in more detail. The present paper summarizes CFD analyses on the impact of the local test site orography on the wake development of a virtual wind turbine. The effects of the orography are identified by comparative simulations for the same turbine using comparative wind situation in flat terrain.

The objective of the present work is to develop a tool able to predict, in a computationally affordable way, the unsteady wind turbine power production and loads as well as its wake dynamics, as a function of the turbine dynamics and incoming wind conditions. Based on the lessons learned from a previous study about the characterization of the unsteady wake dynamics, the framework for an operational wake model is presented. The approach relies on an underlying vorticity-based skeleton consisting of different components, such as a regularized Vortex Sheet Tube (VST) and Vortex Dipole Line (VDL). Physically based evolution equations, accounting for the various flow phenomena occurring in the wake (such as advection, turbulent diffusion/core spreading, source/sink terms, etc.), are then derived. Once calibrated, the wake model is shown to be in good agreement with results of high-fidelity Large Eddy Simulations (LES) obtained using an Immersed Lifting Line-enabled Vortex Particle-Mesh method.

The current trend of the wind energy industry aims for large scale turbines installed in wind farms. This brings a renewed interest in vertical axis wind turbines (VAWTs) since they have several advantages over the traditional Horizontal Axis Wind Tubines (HAWTs) for mitigating the new challenges. However, operating VAWTs are characterized by complex aerodynamics phenomena, presenting considerable challenges for modeling tools. An accurate and reliable simulation tool for predicting the interaction between the obtained wake of an operating VAWT and the flow in atmospheric open sites is fundamental for optimizing the design and location of wind energy facility projects. The present work studies the wake produced by a VAWT and how it is affected by the surface roughness of the terrain, without considering the effects of the ambient turbulence intensity. This study was carried out using an actuator line model (ALM), and it was implemented using the open-source CFD library OpenFOAM to solve the governing equations and to compute the resulting flow fields. An operational H-shaped VAWT model was tested, for which experimental activity has been performed at an open site north of Uppsala-Sweden. Different terrains with similar inflow velocities have been evaluated. Simulated velocity and vorticity of representative sections have been analyzed. Numerical results were validated using normal forces measurements, showing reasonable agreement.

In this wind tunnel campaign, detailed wake measurements behind two different model wind turbines in yawed conditions were performed. The wake deflections were quantified by estimating the rotor-averaged available power within the wake. By using two different model wind turbines, the influence of the rotor design and turbine geometry on the wake deflection caused by a yaw misalignment of 30° could be judged. It was found that the wake deflections three rotor diameters downstream were equal while at six rotor diameters downstream insignificant differences were observed. The results compare well with previous experimental and numerical studies.

The Actuator Line Method exists for more than a decade and has become a well established choice for simulating wind rotors in computational fluid dynamics. Numerous implementations exist and are used in the wind energy research community. These codes were verified by experimental data such as the MEXICO experiment. Often the verification against other codes were made on a very broad scale. Therefore this study attempts first a validation by comparing two different implementations, namely an adapted version of SOWFA/OpenFOAM and EllipSys3D and also a verification by comparing against experimental results from the MEXICO and NEW MEXICO experiments.

The aim of the current investigation is the rotor vibrations generated by the disturbances caused the different types of incoming wake-like flows. Those wakes arriving at the tested rotor were created by two ways: a passive wake generator (immobile disk) and an upstream rotating rotor as an active wake generator. The influence of both wakes on the tested rotor was studied in a water flume. A model of the tested three-bladed rotor designed using Glauert's optimum theory at an optimal tip speed ratio λ = 5 was placed in both "passive" and "active" wakes to recognize dissimilarities on the vibrations of the tested rotor. The distance from the wake generators to the tested rotor was varied from 4 to 8 rotor diameters. Also, the shift between the rotor axis and axis of the incoming wakes was changed to 0, 0.5 and 1 rotor diameters. The flow condition before rotor was measured with high temporal accuracy using LDA. The turbulent intensity of the incoming wake flows changed from 3 to 16% due to the types of the wake generators. Power and thrust characteristics and their pulsations of the tested rotor were measured by strain gauges. The dependences of power coefficients from tip speed ratios and positions of the wake generators were documented. The present study showed a strong influence of the initial flow from the two different wake generators on the rotor vibrations.

The present paper takes a broad view on our previous experimental studies of flows behind different single and dual configurations from passive disks or active rotors to establish new aspects of the wake development [1-4]. The aim of the present examination is to obtain a better understanding of the wake formations and interactions between wind turbines in wind farms. A correlation between independent investigations of the near [1] and far wakes behind single [2] and dual [3-4] systems will be established to the same operating regimes and flow conditions. New examinations of the old data need because two main differences in the wake behaviour for the disk-disk and the rotor-rotor systems were found: the wake intensity grows for the dual disks in comparison with the single one, but in contrast to this, wake intensity behind the dual rotor system is smaller than the one behind a single rotor. These differences may be explained by an influence of the rotor tip vortices which are absent in the disk-disk model. The present retesting of the near and far wake data should provide an evidence of this conclusion.

Wind turbines in a wind power plant experience significant power losses because of aerodynamic interactions between turbines. One control strategy to reduce these losses is known as "wake steering," in which upstream turbines are yawed to direct wakes away from downstream turbines. Previous wake steering research has assumed perfect information, however, there can be significant uncertainty in many aspects of the problem, including wind inflow and various turbine measurements. Uncertainty has significant implications for performance of wake steering strategies. Consequently, the authors formulate and solve an optimization under uncertainty (OUU) problem for finding optimal wake steering strategies in the presence of yaw angle uncertainty. The OUU wake steering strategy is demonstrated on a two-turbine test case and on the utility-scale, offshore Princess Amalia Wind Farm. When we accounted for yaw angle uncertainty in the Princess Amalia Wind Farm case, inflow-direction-specific OUU solutions produced between 0% and 1.4% more power than the deterministically optimized steering strategies, resulting in an overall annual average improvement of 0.2%. More importantly, the deterministic optimization is expected to perform worse and with more downside risk than the OUU result when realistic uncertainty is taken into account. Additionally, the OUU solution produces fewer extreme yaw situations than the deterministic solution.

We present results of the GABLS3 model intercomparison benchmark revisited for wind energy applications. The case consists of a diurnal cycle, measured at the 200-m tall Cabauw tower in the Netherlands, including a nocturnal low-level jet. The benchmark includes a sensitivity analysis of WRF simulations using two input meteorological databases and five planetary boundary-layer schemes. A reference set of mesoscale tendencies is used to drive microscale simulations using RANS k- and LES turbulence models. The validation is based on rotor-based quantities of interest. Cycle-integrated mean absolute errors are used to quantify model performance. The results of the benchmark are used to discuss input uncertainties from mesoscale modelling, different meso-micro coupling strategies (online vs offline) and consistency between RANS and LES codes when dealing with boundary-layer mean flow quantities. Overall, all the microscale simulations produce a consistent coupling with mesoscale forcings.

Each wind turbine experiences a variety of loads during its lifetime, especially inside a wind farm due to the wake effect between the turbines. This paper describes a possibility to observe a load spectrum while considering wake effects in a wind farm by through the turbulence intensity. The turbulence intensity is distributed along the wind rose of Alpha Ventus. For each turbulence intensity, a Weibull characteristic is calculated. The resulting wind fields are used to determine the loads through a multibody simulation of an imaginary wind turbine located at FINO-1, representing a closely placed wind turbine at the outer edge of a wind farm. These loads are analyzed and summed up. As expected, the change of the turbulence intensity due to the wake effect has an impact on the internal loading of a wind turbine inside a wind farm. Based on the assumed loading conditions, the maximum loads increased by a factor of almost 2.5.

For short-term power predictions and estimations of the available power during curtailment of a wind farm, it is necessary to consider the flow dynamics and aerodynamic interactions of the turbines. In this paper, a control-oriented dynamic two-dimensional wind farm model is introduced that aims to incorporate real-time measurements such as flow velocities at turbine locations to estimate the ambient wind farm flow. The model is intended to derive flow predictions for real-time applications. Since fully resolved computational fluid dynamics are too CPU-intensive for such a task, the dynamic model presented in this paper relies on an approximation of the flow equations in a two-dimensional framework. A semi-Lagrangian advection scheme and a step-wise flow solver together offer fast calculation speed, which scales linearly with the number of grid points. In order to emulate effects of realistic three-dimensional wind farm flow, a relaxation of the two-dimensional continuity equation is presented. Furthermore, with little extra computational expense, additional dynamic state variables for various possible applications can be propagated along the wind flow. For instance, a dynamic confidence parameter can provide estimations of the accuracy of flow predictions, while a turbulence parameter adds the possibility to estimate wake induced loads on downstream turbines. In order to demonstrate the performance and validity of the new model it is compared with other models. At first a two turbine reference case is compared with a steady-state model and secondly with results obtained by the dynamic wind farm flow model WFSim. Finally a small wind farm is simulated in order to show the computational scaling of the model.

This paper presents results of a series of numerical simulations in order to study aerodynamic characteristics of the low Reynolds number Selig-Donovan airfoil, SD7003. Large Eddy Simulation (LES) technique is used for all computations at chord-based Reynolds numbers 10,000, 24,000 and 60,000 and simulations have been performed to primarily investigate the role of sub-grid scale (SGS) modeling on the dynamics of flow generated over the airfoil, which has not been dealt with in great detail in the past. It is seen that simulations are increasingly getting influenced by SGS modeling with increasing the Reynolds number, and the effect is visible even at a relatively low chord-Reynolds number of 60,000. Among the tested models, the dynamic Smagorinsky gives the poorest predictions of the flow, with overprediction of lift and a larger separation on airfoils suction side. Among various models, the implicit LES offers closest pressure distribution predictions compared with literature.

In this paper an adaptation of the FLORIS approach is considered that models the wind flow and power production within a wind farm. In preparation to the use of this model for wind farm control, this paper considers the problem of its calibration and validation with the use of experimental observations. The model parameters are first identified based on measurements performed on an isolated scaled wind turbine operated in a boundary layer wind tunnel in various wind-misalignment conditions. Next, the wind farm model is verified with results of experimental tests conducted on three interacting scaled wind turbines. Although some differences in the estimated absolute power are observed, the model appears to be capable of identifying with good accuracy the wind turbine misalignment angles that, by deflecting the wake, lead to maximum power for the investigated layouts.

A comparison between three independent software to estimate the power production and the flow field in a wind farm is conducted, validating them against SCADA (Supervisory, Control And Data Acquisition) data. The three software were ORFEUS, WindSim and WAsP: ORFEUS and WAsP are linearised solvers, while WindSim is fully nonlinear. A wake model (namely a prescribed velocity deficit associated to the turbines) is used by WAsP, while ORFEUS and WindSim use the actuator-disc method to account for the turbines presence. The comparison indicates that ORFEUS and WAsP perform slightly better than WindSim in the assessment of the polar efficiency. The wakes simulated with ORFEUS appear more persistent than the ones of WindSim, which uses a two-equation closure model for the turbulence effects.

Dominant flow structures in the wake region behind the turbine employed in the Blind Test campaign [1], [2] is investigated numerically. The effect on the wake configuration at variable operating conditions are studied. The importance of the introduction of turbine tower inside the numerical framework is highlighted. High-fidelity simulations are performed with Multiple Reference Frame (MRF) numerical methodology. A thorough comparison among the cases is presented, and the wake evolution is analyzed at variable stations downstream of the turbine. Streamlines of flow field traveled towards ground adjacent to turbine tower and strongly dependent on the operating tip speed ratio. Wake is composed of tower shadow superimposed by rotor wake. Shadow of the tower varies from x/R=2 until x/R=4 and breaks down into small vortices with the interaction of rotor wake. This study also shows that the wake distribution consists of two zones; inner zone composed of disturbances generated by blade root, nacelle and the tower, and an outer zone consisting of tip vortices.

Accurate prediction of power generation capability needs proper assessment of blade loading and wake behavior. In this regard, the Sliding Mesh Interface (SMI) approach and the Actuator Line Model (ALM) are two diverse computational fluid dynamics (CFD) based approaches of simulating the turbine behavior, each having its own merits and demerits. The SMI technique simulates the unsteady flow by explicitly modeling the blades and their rotation using a dynamic mesh, while in Actuator Line Model, the blades are not modeled explicitly but each blade is resolved as a rotating line (made of N actuator segments), over which the forces are computed. The current work focuses on simulating an industrial scale reference turbine and in differentiating the near wake dynamics predicted by these two approaches using Large Eddy Simulation (LES) and Proper Orthogonal Decomposition (POD) technique (a data mining tool). Initially, the ALM is compared with FAST model for the prediction of variation of power coefficient with the Tip Speed Ratio (TSR). The ALM is able to capture the varying trend and it predicts a similar optimum tip speed ratio as the FAST model. At this optimum TSR condition, the ALM is compared with the SMI method for a study limited to the near wake region. Comparisons between SMI and ALM shows that : (a) The SMI is predicting more complex 3D nature of the flow, and (b) the POD shows that ALM captures the shear regions of wake but it does not capture the vast compendium of length and time scales of eddies as SMI does. However, despite these limitations, the ALM has been able to capture the qualitative trend in wake deficit and the power coefficient variation with tip speed.

In several papers, the importance of the atmospheric flow in the wake development of wind turbines (WT) has been pointed out, making it clear that it is necessary to have long-term on-field observations for an appropriate description of the wake development, its structure and dynamics. This work presents a statistical approach to wake meandering, y_w, and the relationship that this behavior has with the incoming wind conditions and neighboring wakes. The work was developed in the framework of the French project SMARTEOLE. The study is based on a 7-month measurement campaign in which a pulsed scanning LiDAR system was used. The ground based LiDAR, measures the flow field in a segment such that the wake of two wind turbines can be captured quasi-horizontally. The analysis filters the incoming wind conditions according to the thermal stability, wind direction and wind velocity at hub height; therefore, the wakes that are developed in periods with similar wind conditions are expected to be analogous, hence meandering can be tracked and statistically analyzed. A well-defined wake evolution was found and the uncertainty analysis made on the wake meandering uncovered some interesting characteristics, including the number of samples required to reach a statistical uncertainty on the mean wake position between 2 × 10^-2 D and 8 × 10^-2 D for a confidence interval of 95%.

The effect of a coastline on an offshore wind farm is investigated with a Reynolds-averaged Navier-Stokes (RANS) model. The trends of the RANS model compare relatively well with results from a mesoscale model and measurements of wind turbine power. In addition, challenges of modeling a large domain in RANS are discussed.

The aim of this paper is to present the numerical simulation of waked scaled wind turbines operating in a boundary layer wind tunnel. The simulation uses a LES-lifting-line numerical model. An immersed boundary method in conjunction with an adequate wall model is used to represent the effects of both the wind turbine nacelle and tower, which are shown to have a considerable effect on the wake behavior. Multi-airfoil data calibrated at different Reynolds numbers are used to account for the lift and drag characteristics at the low and varying Reynolds conditions encountered in the experiments. The present study focuses on low turbulence inflow conditions and inflow non-uniformity due to wind tunnel characteristics, while higher turbulence conditions are considered in a separate study. The numerical model is validated by using experimental data obtained during test campaigns conducted with the scaled wind farm facility. The simulation and experimental results are compared in terms of power capture, rotor thrust, downstream velocity profiles and turbulence intensity.

The aim of the present paper is to validate a wind farm LES framework in the context of two distinct wake redirection techniques: yaw misalignment and individual cyclic pitch control. A test campaign was conducted using scaled wind turbine models in a boundary layer wind tunnel, where both particle image velocimetry and hot-wire thermo anemometers were used to obtain high quality measurements of the downstream flow. A LiDAR system was also employed to determine the non-uniformity of the inflow velocity field. A high-fidelity large-eddy simulation lifting-line model was used to simulate the aerodynamic behavior of the system, including the geometry of the wind turbine nacelle and tower. A tuning-free Lagrangian scale-dependent dynamic approach was adopted to improve the sub-grid scale modeling. Comparisons with experimental measurements are used to systematically validate the simulations. The LES results are in good agreement with the PIV and hot-wire data in terms of time-averaged wake profiles, turbulence intensity and Reynolds shear stresses. Discrepancies are also highlighted, to guide future improvements.

In the present paper the actuator line method is compared with fully resolved wind turbine simulations in offshore and complex terrain applications. In such flow fields, which are characterized by non-homogeneous and unsteady velocity distributions in the rotor plane, unsteady aerodynamic effects are likely and it is unclear how these characterize the wake development and load behavior of the wind turbine. The wake properties and loads are therefore compared for the case of a 5 MW wind turbine operating in a typical maritime atmosphere and a 2.4 MW onshore turbine located at a complex terrain site downstream of an escarpment. It was found that the actuator line predicts the wake structure, wake deflection and wake deficit in good agreement with the fully resolved simulation. However, an overestimation of velocity fluctuations was observed.

This work aims to reproduce the measured atmospheric conditions during one day of the CWEX-11 campaign, with a transient LES. The selected period includes several interesting atmospheric conditions for wind power generation such as a nocturnal low-level jet, a highly turbulent convective daytime boundary layer, as well as a distinct evening transition between daytime and nocturnal boundary layers. To include synoptic conditions, large-scale forcing profiles for the LES were derived from a mesoscale simulation with the WRF model. A comparison with lidar measurements shows that the trend of the wind conditions and the diurnal cycle is well replicated by the model chain. Selected periods of the day are simulated with the NREL 5MW turbine model, followed by a qualitative comparison of measured and simulated wakes. We find a strong dependency of the meandering and the shape of the wake on wind profile and turbulence, while a categorization by Obukhov length is less representative for the different conditions. As the veer in the wind profile increases, the deviation of the wind direction at hub height from the direction of the largest wake impact also increases.

In this paper we further investigate and validate the novel theoretical model of very large wind farms proposed recently by Nishino (J. Phys.: Conf. Ser. 753, 032054, 2016). One of the key features of the Nishino model is that a theoretically optimal turbine resistance (as well as optimal 'turbine-scale' and 'farm-scale' wind speed reduction rates) can be predicted analytically as a function of the farm density and the natural bottom friction observed before constructing the farm. To validate this theoretical model, a new set of 3D Reynolds-averaged Navier-Stokes (RANS) simulations are performed of a fully developed wind farm boundary layer over an aligned and staggered array of actuator discs with various disc resistance, inter-disc spacing and bottom roughness values. The results show that the theoretical model, which employs only one empirical model parameter, can be easily calibrated to predict very well the performance of various staggered arrays of actuator discs. This suggests the usefulness of the theoretical model not only for providing an upper limit to the performance of ideal large arrays but also for predicting the performance of realistic large arrays. The results also highlight the important fact that the optimal turbine resistance can be significantly smaller in a dense wind farm than in a sparse wind farm.

The present work studies the large coherent structures in large eddy simulations of windfarms using proper orthogonal decomposition (POD) method. In order to evaluate the effect of wind turbines on coherent structures, we consider three cases. One is a reference flow of a neutral atmospheric boundary layer and the other two are periodic and developing aligned windfarms. The number of wind turbines is large, 16 × 12 for periodic windfarm, and 12 × 12 for developing windfarm. The simulations are run for a long time in order to generate a sufficient database for POD analysis. In all cases, elongated streamwise counter rotating roll structures, covering 1 or 2 turbines in spanwise direction, are identified as the dominant POD mode. Another pattern, varying in streamwise direction, also appears in all the three cases.