Table of contents

The European Academy of Wind Energy (eawe) was pleased to announce its 4th scientific conference The Science of Making Torque from Wind. Predecessors have successfully been arranged in Delft, The Netherlands (2004), Lyngby, Denmark (2007) and Heraklion, Greece (2010). During the years the Torque Conference has established itself as Europe's leading scientific wind energy conference. The 2012 edition had been organized in the same tradition. More than 300 experts from academia and industry discussed the latest results and developments in fundamental and applied wind energy research, making this Science of Making Torque from Wind conference the largest one to that date.

The seven keynote lectures provided the delegates with a unique overview on the state-of-the-art of science and technology. In over twenty sessions the participants discussed the most recent results in wind energy research. From numerical models to sophisticated experiments, from flow optimizations to structural designs, the numerous presentations covered a huge spectrum of ongoing scientific activities.

The proceedings of the Torque 2012 combine the 110 papers that have passed the review process. We would like to thank all those who have been involved in organizing the conference and putting together these proceedings, including keynote speakers, session chairs and the enormous amount of reviewers involved. We are especially grateful to Gijs van Kuik for his untiring support. We also deeply appreciate the logistical support and technical services of the University of Oldenburg and the financial support of the State of Lower Saxony. At IOP we would like to thank Anete Ashton for her continuous encouraging support.

We are looking forward to all future Torque Conferences, offering an excellent platform for the exchange of the latest and greatest scientific developments in the field of wind energy.

Oldenburg, Germany, October 2014

Elke Seidel, Detlev Heinemann, Martin Kühn, Joachim Peinke and Stephan Barth ForWind – University of Oldenburg

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.

Continued inquiry into rotor and blade aerodynamics remains crucial for achieving accurate, reliable prediction of wind turbine power performance under yawed conditions. To exploit key advantages conferred by controlled inflow conditions, EU-JOULE DATA Project and UAE Phase VI experimental data were used to characterize rotor power production under yawed conditions. Anomalies in rotor power variation with yaw error were observed, and the underlying fluid dynamic interactions were isolated. Unlike currently recognized influences caused by angled inflow and skewed wake, which may be considered potential flow interactions, these anomalies were linked to pronounced viscous and unsteady effects.

In this study, the effect of the parameters playing a role in the root flow behavior of HAWT are only partly understood. To better reveal the root flow properties, this study presents the progression of HAWT blade root flow at two different blade geometries and at two different tip speed ratios. The effects of the geometry and the tip speed ratio on the root flow behavior and on the evolution of the root flow features are investigated. This study aims to answer the following questions: (i) What are the effects of the blade geometry and tip speed ratio on the root flow behavior? (ii) How are the blade wake and the root vortex evolution affected by the change of these parameters? The analysis of the velocity fields shows that the radial flow behavior changes with different blade geometries but a remarkable difference in the radial flow behavior is not observed with the change of tip speed ratio. The formation of the wake is different at three test cases because of different loading that the blades are encountered. From the circulation distribution along the blades, while a strong root vortex can be observed in Blade 1, the bound vorticity along Blade 2 builds up gradually when moving outboard, and do not show a trace of a strong root vortex.

Schmitz and Blade Element Momentum (BEM) theories are integrated to a gradient based optimization algorithm to optimize the blade shape of a horizontal axis wind turbine (HAWT). The Schmitz theory is used to generate an initial blade design. BEM theory is used to calculate the forces, torque and power extracted by the turbine. The airfoil shape (NREL S809) is kept the same, so that the shape optimization comprises only the chord and the pitch angle distribution. The gradient based optimization of the blade shape is constrained to the torque-rotational speed characteristic of the generator, which is going to be a part of the experimental set-up used to validate the results of the optimization study. Hence, the objective of the optimization is the maximization of the turbines power coefficient C_p while keeping the torque matched to that of the generator. The wind velocities and the rotational speeds are limited to those achievable in the wind tunnel and by the generator, respectively. After finding the optimum blade shape with the maximum C_p within the given range of parameters, the C_p of the turbine is evaluated at wind-speeds deviating from the optimum operating condition. For this purpose, a second optimization algorithm is used to find out the correct rotational speed for a given wind-speed, which is again constrained to the generator's torque rotational speed characteristic. The design and optimization procedures are later validated by high-fidelity numerical simulations. The agreement between the design and the numerical simulations is very satisfactory.

Atmospheric stability is known to influence wind farm power output, by affecting power losses due to wakes. This research tries to answer what atmospheric stability does to the power production and how conventional simulations using the Jensen wake model compare and can be improved. Data is used from two offshore wind farms, Egmond aan Zee (OWEZ) and North Hoyle. Stability distributions are determined using metmast data. By combining this data with the production data, the influence of stability on the power output is studied. It is found that very unstable conditions result in higher power output (i.e. smaller wake losses) than near-neutral conditions, and these again show higher power output than during very stable conditions. Differences in normalized power output of 10-20% exist between the very unstable and very stable conditions. Simulations can be improved by adapting the wake decay constant (WDC). Observed WDC values are k ≥ TI, as opposed to the conventional k ≈ 0.5TI. A hypothesis for further research is proposed regarding the influence of vertical turbulence.

The turbulence in the interior of an idealised wind farm is simulated using Large Eddy Simulation and the Actuator Line technique implemented in the Navier-Stokes equations. The simulation is carried out for an 'infinitely' long row of turbines simulated by applying cyclic boundary conditions at the inlet and outlet. The simulations investigate the turbulence inherent to the wind turbines as no ambient turbulence or shear is added to this idealised case. A Reduced Order Model for the highly turbulent flow deep inside a wind farm is proposed based on a Proper Orthogonal Decomposition. The reconstructed flow is shown to capture the large scale motions of the highly turbulent flow.

This paper presents an aerodynamic and aeroelastic analysis of the MEXICO wind turbine, using the compressible HMB solver of Liverpool. The aeroelasticity of the blade, as well as the effect of a low-Mach scheme were studied for the zero-yaw 15m/s wind case and steady- state computations. The wake developed behind the rotor was also extracted and compared with the experimental data, using the compressible solver and a low-Mach scheme.

It was found that the loads were not sensitive to the Mach number effects, although the low-Mach scheme improved the wake predictions. The sensitivity of the results to the blade structural properties was also highlighted.

The paper presents an analysis of the aeroelastic loads on a wind turbine in normal operation. The characteristic of the loads causing the highest fatigue damage are identified, so to provide indications to the development of active load alleviation systems for smart- rotor applications. Fatigue analysis is performed using rain-flow counting and Palmgren-Miner linear damage assumption; the contribution to life-time fatigue damage from deterministic load variations is quantified, as well as the contributions from operation at different mean wind speeds. A method is proposed to retrieve an estimation of the load frequencies yielding the highest fatigue contributions from the bending moment spectra. The results are in good agreement with rain-flow counting analysis on filtered time series, and, for the blade loads, show dominant contributions from frequencies close to the rotational one; negligible fatigue contributions are reported for loads with frequencies above 2 Hz.

During the last decade research on the field of smart rotor has advanced significantly. Fundamental aerodynamics, structural and control concepts have been established and simulators created for distributed flaps on wind turbine blades, which are considered the most promising option. Also a proof of concept has been done under laboratory conditions. However, the results obtained under these conditions can only be partially transfer to the real application as the control authority of smart rotors is limited compared to full pitch control. The steps that need to be taken before smart rotors can be successfully exploited are in the design of reliable systems that can operate under environmental conditions without inspections. Besides that, other potential advantages of distributed control need to be established such as the effect on other components of a wind turbine for example the gear box or the power system. Finally, it is necessary to investigate what benefits can be achieved if blades are designed with distributed control right from the start instead of applying control schemes to already existing turbines.

With the help of a local stability analysis the coefficient range of a discrete damper, used for centrifugal forced, mechanical pitch system of small wind turbines (SWT), is gained for equilibrium points. – By a global stability analysis the gained coefficient range can be validated. An appropriate approach by Takagi-Sugeno is presented in the paper.

A quasi-simultaneous interaction method is applied to predict 2D and 3D aerodynamic flows. This method is suitable for offshore wind turbine design software as it is a very accurate and computationally reasonably cheap method.

This study shows the results for a NACA 0012 airfoil. The two applied solvers converge to the experimental values when the grid is refined. We also show that in separation the eigenvalues remain positive thus avoiding the Goldstein singularity at separation.

In 3D we show a flow over a dent in which separation occurs. A rotating flat plat is used to show the applicability of the method for rotating flows.

The shown capabilities of the method indicate that the quasi-simultaneous interaction method is suitable for design methods for offshore wind turbine blades.

In recent years there has been much interest in the possible use of LIDAR systems for improving the performance of wind turbine controllers, by providing preview information about the approaching wind field. Various potential benefits have been suggested, and experimental measurements have sometimes been used to claim surprising gains in performance. This paper reports on an independent study which has used detailed analytical methods for two main purposes: firstly to try to evaluate the likely benefits of LIDAR-assisted control objectively, and secondly to provide advice to LIDAR manufacturers about the characteristics of LIDAR systems which are most likely to be of value for this application. Many different LIDAR configurations were compared: as a general conclusion, systems should be able to sample at least 10 points every second, reasonably distributed around the swept area, and allowing a look-ahead time of a few seconds. An important conclusion is that the main benefit of the LIDAR will be to enhance of collective pitch control to reduce thrust-related fatigue loads; there is some indication that extreme loads can also be reduced, but this depends on other considerations which are discussed in the paper. LIDAR-assisted individual pitch control, optimal C_p tracking and yaw control were also investigated, but the benefits over conventional methods are less clear.

A new analytical formulation of the tip-loss factor is established based on helical vortex filament solutions. The derived tip-loss factor can be applied to wind-turbines, propellers or other rotary wings. Similar numerical formulations are used to assess the influence of wake expansion on tip-losses. Theodorsen's theory is successfully applied for the first time to assess the wake expansion behind a wind turbine. The tip-loss corrections obtained are compared with the ones from Prandtl and Glauert and implemented within a new Blade Element Momentum(BEM) code. Wake expansion is seen to reduce tip-losses and have a greater influence than wake distortion.

The drive to upscale offshore wind turbines relates especially to possiblereductions in O&M and electrical interconnection costs per MW of installed capacity.Even with best current technologies, designs with rated capacity above about 3 MW are less cost effective exfactory per rated MW(turbine system costs) than smaller machines.Very large offshore wind turbines are thereforejustifiedprimarily by overall offshore project economics. Furthermore, continuing progress in materials and structures has been essential to avoid severe penalties in the power/mass ratio of large multi-MW machines.The multi-rotor concept employs many small rotors to maximise energy capture area withminimum systemvolume. Previous work has indicated that this can enablea very large reduction in the total weight and cost of rotors and drive trains compared to an equivalent large single rotor system.Thus the multi rotor concept may enable rated capacities of 20 MW or more at a single maintenancesite.

Establishing the cost benefit of a multi rotor system requires examination of solutions for the support structure and yawing, ensuring aerodynamic losses from rotor interaction are not significant and that overall logistics, with much increased part count (more reliable components) and less consequence of single failuresare favourable. This paper addresses the viability of a support structure in respect of structural concept and likely weight as one necessary step in exploring the potential of the multi rotor concept.

The effect of a downstream turbine on the production of a turbine located upstream of the latter is studied in this work. This is done through the use of two CFD simulation codes, namely OpenFOAM and EllipSys3D, which solve the Navier-Stokes equations in their incompressible form using a finite volume approach. In both EllipSys3D and OpenFoam, the LES (Large Eddy Simulation) technique is used for modelling turbulence. The wind turbine rotors are modelled as actuator disks whose loading is determined through the use of tabulated airfoil data by applying the blade-element method. A generator torque controller is used in both simulation methods to ensure that the simulated turbines adapt, in terms of rotational velocity, to the inflow conditions they are submited to. Results from both simulation codes, although they differ slightly, show that the downstream turbine affects the upstream one when the spacing between the turbines is small. This is also suggested to be the case looking at measurements performed at the Lillgrund offshore wind farm, whose turbines are located unusually close to each other. However, for distances used in today's typical wind farms, this effect is shown by our calculations not to be significant.

High-quality computer simulations are required when designing floating wind turbines because of the complex dynamic responses that are inherent with a high number of degrees of freedom and variable metocean conditions. In 2007, the FAST wind turbine simulation tool, developed and maintained by the U.S. Department of Energy's (DOE's) National Renewable Energy Laboratory (NREL), was expanded to include capabilities that are suitable for modeling floating offshore wind turbines. In an effort to validate FAST and other offshore wind energy modeling tools, DOE funded the DeepCwind project that tested three prototype floating wind turbines at 1/50^th scale in a wave basin, including a semisubmersible, a tension-leg platform, and a spar buoy. This paper describes the use of the results of the spar wave basin tests to calibrate and validate the FAST offshore floating simulation tool, and presents some initial results of simulated dynamic responses of the spar to several combinations of wind and sea states. Wave basin tests with the spar attached to a scale model of the NREL 5-megawatt reference wind turbine were performed at the Maritime Research Institute Netherlands under the DeepCwind project. This project included free-decay tests, tests with steady or turbulent wind and still water (both periodic and irregular waves with no wind), and combined wind/wave tests. The resulting data from the 1/50^th model was scaled using Froude scaling to full size and used to calibrate and validate a full-size simulated model in FAST. Results of the model calibration and validation include successes, subtleties, and limitations of both wave basin testing and FAST modeling capabilities.

This paper deals with the performance modelling and analysis of offshore wind turbine-driven hydraulic pumps. The concept consists of an open loop hydraulic system with the rotor main shaft directly coupled to a swash plate pump to supply pressurised sea water. A mathematical model is derived to cater for the steady state behaviour of entire system. A simplified model for the pump is implemented together with different control scheme options for regulating the rotor shaft power. A new control scheme is investigated, based on the combined use of hydraulic pressure and pitch control. Using a steady-state analysis, the study shows how the adoption of alternative control schemes in a the wind turbine-hydraulic pump system may result in higher energy yields than those from a conventional system with an electrical generator and standard pitch control for power regulation. This is in particular the case with the new control scheme investigated in this study that is based on the combined use of pressure and rotor blade pitch control.

The cross-spectral phases between velocity components at two heights are analyzed from observations at the Høvsøre test site under diabatic conditions. These phases represent the degree to which turbulence sensed at one height leads (or lags) in time the turbulence sensed at the other height. The phase angle of the cross-wind component is observed to be significantly greater than the phase for the along-wind component, which in turn is greater than the phase for the vertical component. The cross-wind and along-wind phases increase with stream-wise wavenumber and vertical separation distance, but there is no significant change in the phase angle of vertical velocity. The phase angles for all atmospheric stabilities show similar order in phasing. The phase angles from the Høvsøre observations under neutral condition are compared with a rapid distortion theory model which show similar order in phase shift.

Reaching for higher wind resources beyond the water depth limitations of monopile wind turbines, there has arisen the alternative of using floating wind turbines. But the response of wave induced loads significantly increases for floating wind turbines. Applying conventional onshore control strategies to floating wind turbines has been shown to impose negative damped oscillations in fore-aft due to the low natural frequency of the floating structure. Thus, we suggest a control loop extension of the onshore controller which stabilizes the system and reduces the wave disturbance. The results shows that when adding the suggested control loop with disturbance reduction to the system, improved performance is observed in power fluctuations, blade pitch activity, and platform oscillations.

Two active aerodynamic load control (AALC) devices coupled with a control algorithm are shown to decrease the change in lift force experienced by an airfoil during a change in freestream velocity. Microtabs are small (1% chord) surfaces deployed perpendicular to an airfoil, while microjets are pneumatic jets with flow perpendicular to the surface of the airfoil near the trailing edge. Both devices are capable of producing a rapid change in an airfoil's lift coefficient. A control algorithm for microtabs has been tested in a wind tunnel using a modified S819 airfoil, and a microjet control algorithm has been simulated for a NACA 0012 airfoil using OVERFLOW. In both cases, the AALC devices have shown the ability to mitigate the changes in lift during a gust.

A relation between Cost Of Energy, COE, maximum allowed tip speed, and rated wind speed, is obtained for wind turbines with a given goal rated power. The wind regime is characterised by the corresponding parameters of the probability density function of wind speed. The non-dimensional characteristics of the rotor: number of blades, the blade radial distributions of local solidity, twist, angle, and airfoil type, play the role of parameters in the mentioned relation. The COE is estimated using a cost model commonly used by the designers. This cost model requires basic design data such as the rotor radius and the ratio between the hub height and the rotor radius. Certain design options, DO, related to the technology of the power plant, tower and blades are also required as inputs. The function obtained for the COE can be explored to find those values of rotor radius that give rise to minimum cost of energy for a given wind regime as the tip speed limitation changes. The analysis reveals that iso-COE lines evolve parallel to iso-radius lines for large values of limit tip speed but that this is not the case for small values of the tip speed limits. It is concluded that., as the tip speed limit decreases, the optimum decision for keeping minimum COE values can be: a) reducing the rotor radius for places with high weibull scale parameter or b) increasing the rotor radius for places with low weibull scale parameter.

The flow topology on two scaled models (1:230 and 1:115) of the Bolund Island is analysed in two wind tunnels, focusing on the characteristics of the detachment pattern when the wind blows from 270° wind direction and the atmospheric condition is neutral. Since the experiments are designed as the simplest possible reference cases, no additional roughness is added neither to the models surface nor to the wind tunnel floor. Pressure measurements on the surface of the 1:230 scale model are used to estimate the horizontal extension of the intermittent recirculation region, by applying the diagnostic means based in exploring the pressure statistics, proposed in the literature for characterising bubbles on canonical obstacles. The analysis is done for a range of Reynolds numbers based on the mean undisturbed wind speed, U_∞ and the maximum height of the island, h[5.1×10⁴,8.5×10⁴]. An isoheight mapping of the velocity field is obtained using 3D hotwire (3D HW). The velocity field in a vertical plane is determined using 3D HW and 2D particle image velocimetry (PIV) on the 1:115 scale model in order to reproduce and complete already existing results in the literature.

The values of the tip speed ratio and blade pitch angle that yield maximum power coefficient are calculated for a rotor operating in yawed conditions. In a first step, the power coefficient is determined using a model based on the blade element momentum theory (BEMT) which includes a Prandtl-Glauert root-tip losses correction, a non-uniform model for the axial and tangential induction factors, and a model of the rotational augmentation effects. The BEMT model is validated with the experimental data from the NREL-UAE. The maximum values of the power coefficient are determined for different yaw angles and the corresponding values of the tip speed ratio and blade control angle are obtained. The maximum power coefficient using these optimum laws is compared to the maximum power coefficient using the optimum laws of the non-yawed case and it is shown that there is a gain in the power coefficient. For the case study presented in this paper it has been found that for yaw angles of 30° about 10% of the power coefficient can be recovered.

A model wind turbine has been designed at the University of Hamburg within the scope of the FP7 fundend project WAUDIT. The purpose of the experiment described in this paper is to characterize the performances of two rotors by means of measuring the thrust coefficient C_t. C_t is a similarity parameter for the wake and is thought to be the most effective one. Its value has been directly measured using a force balance and indirectly calculated from the velocity profiles measured three diameters downstream of the rotor with hot wire anemometry. Results show that, in order to reproduce the wake behaviour, the matching of the C_t, which is a quantitative achievement, has to be integrated with measurements such as velocity profiles in the wake. In fact the velocity deficit illustrates the mechanism of transforming the axial momentum into torque assuring qualitatively the proper reproduction of the wake. This latter information assures that the achievement of a certain thrust force acting on the rotor is due to its performances in transforming the axial momentum into torque and not an effect of other phenomena such as a stall at the blades.

This paper presents two H_∞ multivariable robust controllers based on blade root sensors' information for individual pitch angle control. The wind turbine of 5 MW defined in the Upwind European project is the reference non-linear model used in this research work, which has been modelled in the GH Bladed 4.0 software package. The main objective of these controllers is load mitigation in different components of wind turbines during power production in the above rated control zone. The first proposed multi-input multi-output (MIMO) individual pitch H" controller mitigates the wind effect on the tower side-to-side acceleration and reduces the asymmetrical loads which appear in the rotor due to its misalignment. The second individual pitch H" multivariable controller mitigates the loads on the three blades reducing the wind effect on the bending flapwise and edgewise momentums in the blades. The designed H" controllers have been validated in GH Bladed and an exhaustive analysis has been carried out to calculate fatigue load reduction on wind turbine components, as well as to analyze load mitigation in some extreme cases.

In order to evaluate aerodynamic loads on floating offshore wind turbines, advanced dynamic analysis tools are required. As a unified model that can represent both dynamic inflow and skewed inflow effects in it basic formulation, a wake model based on a vortex ring formulation is discussed. Such a model presents a good intermediate solution between computationally efficient but simple momentum balance methods and computationally expensive but complete computational fluid dynamics models. The model introduced is shown to be capable of modelling typical steady and unsteady test cases with reasonable accuracy.

The reduction of wind turbine blade loads is an important issue in the reduction of the costs of energy production. Reduction of the loads of a non-cyclic nature requires so-called smart rotor control, which involves the application of distributed actuators and sensors to provide fast and local changes in aerodynamic performance. This paper investigates the use of synthetic jets for smart rotor control. Synthetic jets are formed by ingesting low-momentum fluid from the boundary layer along the blade into a cavity and subsequently ejecting this fluid with a higher momentum. We focus on the observed flow phenomena and the ability to use these to obtain the desired changes of the aerodynamic properties of a blade section. To this end, numerical simulations and wind tunnel experiments of synthetic jet actuation on a non-rotating NACA0018 airfoil have been performed. The synthetic jets are long spanwise slits, located close to the trailing edge and directed perpendicularly to the surface of the airfoil. Due to limitations of the present experimental setup in terms of performance of the synthetic jets, the main focus is on the numerical flow simulations. The present results show that high-frequency synthetic jet actuation close to the trailing edge can induce changes in the effective angle of attack up to approximately 2.9°.

Data from four well instrumented met masts located in heavily forested European sites in different locations and terrain types are examined. Seven stability metrics are applied to the data sets and a novel method is used to identify the metric which most consistently identifies stability events of importance for wind energy generation. It was found that the Obukhov length, as calculated by fast response sonic anemometer, provides the most reliable results in these highly complex sites. It was also found that non-neutral stabilities can be expected a significant portion of the time for wind speeds of less than 10 m/s at the considered sites.

In this paper the results are presented of experiments to prove an innovative concept for passive torque control of variable speed wind turbines using fluid power technology. It is demonstrated that by correctly configuring the hydraulic drive train, the wind turbine rotor operates at or near maximum aerodynamic efficiency for below rated wind speeds. The experiments with a small horizontal-axis wind turbine rotor, coupled to a hydraulic circuit, were conducted at the Open Jet Facility of the Delft University of Technology. In theory, the placement of a nozzle at the end of the hydraulic circuit causes the pressure and hence the rotor torque to increase quadratically with flow speed and hence rotation speed. The rotor torque is limited by a pressure relief valve. Results from the experiments proved the functionality of this passive speed control concept. By selecting the correct nozzle outlet area the rotor operates at or near the optimum tip speed ratio.

Power curves for offshore wind turbines within the German offshore wind farm alpha ventus were derived based on the IEC standard. Binning in groups of shear and turbulence intensity as measures of atmospheric stability were performed. The derived power curves show a strong dependency on these two parameters. Differences of up to 15% in power output between unstable and stable stratification in the non-wake case occur. For wind turbines within the wake of others the effects are even more pronounced. Here, the differences in power production between the stability classes approach 20%. This dependency of the power curves on stability can cause significant miscalculations of instantaneous power production, long-term energy yield and loads. Parameters other than the hub height wind speed are often not taken into account in state-of-the-art wind power forecasts. This can lead to substantial over- or underestimation of the resulting power.

The tower is one of the major components in wind turbines with a contribution to the cost of energy of 8 to 12% [1]. In this overview the load situation of the tower will be described in terms of sources of loads, load components and fatigue contribution. Then two load reduction control schemes are described along with simulation and field test results. Pitch Balancing is described as a method to reduce aerodynamic asymmetry and the resulting fatigue loads. Active Tower Damping is reducing the tower oscillations by applying appropiate pitch angle changes. A field test was conducted on an Areva M5000 wind turbine.

The main objective of this research work is to develop and evaluate several coupling methods between operational Numerical Weather Prediction (NWP) model and Computational Fluid Dynamics (CFD) model and data assimilate the field measurements into the CFD model. To address the problem of high spatial variation of the topography on the domain lateral boundaries between NWP and CFD domain boundaries, 3 methods – translation, extrapolation and Cressman interpolation are used to impose the NWP model data on the CFD domain lateral boundaries. Newtonian relaxation data assimilation technique is used to incorporate the field measurement data into the CFD simulations. These techniques are studied in a complex site located in southern France. Comparison of wind profiles between the CFD simulation, measurements and CFD simulation with data assimilation are discussed. This combination of state-of-the-art techniques in NWP, CFD, and field data assimilation will provide the basis of a more accurate wind resource assessment method.

The wake recovery behind the Horns Rev wind farm is analysed to investigate the applicability of Large Eddy Simulations (LES) in combination with an actuator disc method (ACD) for farm to farm interaction studies. Periodic boundary conditions on the lateral boundaries are used to model the wind farm (as infinitely wide), using only two columns of turbines. The meteorological conditions of the site are taken into account by introducing wind shear and pre-generated synthetic turbulence to the simulation domain using body forces. Simulations are carried out to study the power production and the velocity deficit in the farm wake. The results are compared to the actual power production as well as to wind measurements at 2 km and 6 km behind the wind farm. The simulated power production inside the farm shows an overall good correlation with the real production, but is slightly overpredicted in the most downstream rows. The simulations overpredict the wake recovery, namely the wind velocity, at long distances behind the farm. Further studies are needed before the presented method can be applied for the simulation of long distance wakes. Suggested parameters to be studied are the development of the turbulence downstream in the domain and the impact of the grid resolution.

Offshore wind turbines operate in a complex unsteady flow environment which causes unsteady aerodynamic loads. The unsteady flow environment is characterized by a high degree of uncertainty. In addition, geometry variations and material imperfections also cause uncertainties in the design process. Probabilistic design methods consider these uncertainties in order to reach acceptable reliability and safety levels for offshore wind turbines. Variations of the rotor blade geometry influence the aerodynamic loads which also affect the reliability of other wind turbine components. Therefore, the present paper is dealing with geometric uncertainties of the rotor blades. These can arise from manufacturing tolerances and operational wear of the blades. First, the effect of geometry variations of wind turbine airfoils on the lift and drag coefficients are investigated using a Latin hypercube sampling. Then, the resulting effects on the performance and the blade loads of an offshore wind turbine are analyzed. The variations of the airfoil geometry lead to a significant scatter of the lift and drag coefficients which also affects the damage-equivalent flapwise bending moments. In contrast to that, the effects on the power and the annual energy production are almost negligible with regard to the assumptions made.

In this paper, we investigate the characteristics of a wind turbine under blade pitch angle and shaft speed sensor faults as well as pitch actuator faults. A land-based NREL 5MW variable speed pitch reg- ulated wind turbine is considered as a reference. The conventional collective blade pitch angle controller strategy with independent pitch actuators control is used for load reduction. The wind turbine class is IEC-BII. The main purpose is to investigate the severity of end effects on structural loads and responses and consequently identify the high-risk components according to the type and amplitude of fault using a servo-aero-elastic simulation code, HAWC2. Both transient and steady state effects of faults are studied. Such information is useful for wind turbine fault detection and identification as well as system reliability analysis. Results show the effects of faults on wind turbine power output and responses. Pitch sensor faults mainly affects the vibration of shaft main bearing, while generator power and aerodynamic thrust are not changed significantly, due to independent pitch actuator control of three blades. Shaft speed sensor faults can seriously affect the generator power and aerodynamic thrust. Pitch actuator faults can result in fully pitching of the blade, and consequently rotor stops due to negative aerodynamic torque.

Due to the increasing penetration of wind energy into power systems, it becomes critical to reduce the impact of wind energy on the stability and reliability of the overall power system. In precedent works, Shen and his co-workers developed a re-designed operation schema to run wind turbines in strong wind conditions based on optimization method and standard PI feedback control, which can prevent the typical shutdowns of wind turbines when reaching the cut-out wind speed. In this paper, a new control strategy combing the standard PI feedback control with feedforward controls using the optimization results is investigated for the operation of variable-speed pitch-regulated wind turbines in strong wind conditions. It is shown that the developed control strategy is capable of smoothening the power output of wind turbine and avoiding its sudden showdown at high wind speeds without worsening the loads on rotor and blades.

According to experimental observations, the vortices generated by vortex generators have previously been observed to be self-similar for both the axial (u_z) and azimuthal (u_θ) velocity profiles. Further, the measured vortices have been observed to obey the criteria for helical symmetry. This is a powerful result, since it reduces the highly complex flow to merely four parameters. In the present work, corresponding computer simulations using Reynolds- Averaged Navier-Stokes equations have been carried out and compared to the experimental observations. The main objective of this study is to investigate how well the simulations can reproduce the physics of the flow and if the same analytical model can be applied. Using this model, parametric studies can be significantly reduced and, further, reliable simulations can substantially reduce the costs of the parametric studies themselves.

The maximum fatigue load reduction potential when using trailing edge flaps on mega-watt wind turbines was explored. For this purpose an ideal feed forward control algorithm using the relative velocity and angle of attack at the blade to control the loads was implemented. The algorithm was applied to time series from computations with the aeroelastic code HAWC2 and to measured time series. The fatigue loads could be reduced by 36% in the computations if the inflow sensor was at the same position as the blade load. The decrease of the load reduction potential when the sensor was at a distance from the blade load location was investigated. When the algorithm was applied to measured time series a load reduction of 23% was achieved which is still promissing, but significantly lower than the value achieved in computations.

We have carried out in-depth analyses of boundary-layer wind velocity data within a universal multifractal (UM) framework. Within the UM framework the statistics of a given field are characterised with the help of the three parameters α, C₁ and H. With these three parameters one fully describes the wind velocity fields up to and including the order q_D after which the divergence of statistical moments intervenes. Studies at different sites have shown that the parameter α – the multifractality index – of the horizontal and vertical shears of the horizontal wind remains fairly constant at approximately 1.7. In this study we show how the two remaining parameters C₁ and H vary for two very different sites/datasets and discuss what the consequences of this variability are for the fluctuations of the torque.

From extensive application over a number of years, it has been established that the nonlinear rotor aerodynamics of typical medium and large wind turbines exhibit an effectively global separability property, in other words the aerodynamic torque of the machine can be defined by two independent additive functions. Two versions of the separability of aerodynamic torque for variable speed wind turbines are investigated here; the separated function, related to wind speed, in the first version is only dependent on that variable and not rotor speed and in the second version is only dependent on tip speed ratio. Both provide very good approximations to the aerodynamic torque over extensive neighbourhoods of T₀, at least from 0 to 2T₀.

In order to implement accurate models for wind power ramp forecasting, ramps need to be previously characterised. This issue has been typically addressed by performing binary ramp/non-ramp classifications based on ad-hoc assessed thresholds. However, recent works question this approach. This paper presents the ramp function, an innovative wavelet- based tool which detects and characterises ramp events in wind power time series. The underlying idea is to assess a continuous index related to the ramp intensity at each time step, which is obtained by considering large power output gradients evaluated under different time scales (up to typical ramp durations). The ramp function overcomes some of the drawbacks shown by the aforementioned binary classification and permits forecasters to easily reveal specific features of the ramp behaviour observed at a wind farm. As an example, the daily profile of the ramp-up and ramp-down intensities are obtained for the case of a wind farm located in Spain.

A computational model for predicting the aerodynamic behavior of wind turbine airfoils under rotation and subjected to steady and unsteady motions developed in [1] is presented herein. The model is based on a viscous-inviscid interaction technique using strong coupling between the viscous and inviscid parts. The rotational effects generated by centrifugal and Coriolis forces are introduced in Q³UIC via the streamwise and spanwise integral boundary layer momentum equations. A special inviscid version of the code has been developed to cope with massive separation. To check the ability of the code wind turbine airfoils in steady and unsteady conditions for a large range of angles of attack are considered here. Further, the new quasi-3D code Q³UIC is used to perform a parametric study of a wind turbine airfoil under rotation confined to its boundary layer.

Based on a reduced-order, dynamic nonlinear wind turbine model in Takagi- Sugeno (TS) model structure, a TS state observer is designed as a disturbance observer to estimate the unknown effective wind speed. The TS observer model is an exact representation of the underlying nonlinear model, obtained by means of the sector-nonlinearity approach. The observer gain matrices are obtained by means of a linear matrix inequality (LMI) design approach for optimal fuzzy control, where weighting matrices for the individual system states and outputs are included. The observer is tested in simulations with the aero-elastic code FAST for the NREL 5 MW reference turbine, where it shows a stable behaviour in turbulent wind simulations.

Offshore resource assessment with lidars on floating platforms is a flexible and particularly cost-effective alternative to the conventional meteorological mast solution, that is considered as onshore state-of-the-art transferred to offshore sites, and may enable better and more complete wind resource assessments for the growing offshore wind sector. Wind lidar technology, and remote sensing in general, has already been proven to be a very promising technology for resource assessment and power performance testing onshore. For offshore applications and on floating platforms in particular, the motions from the floating base have to be considered in addition, affecting the wind measurements significantly and causing systematic measurement errors.

We have studied the motions and the corresponding influences on lidar measurements generated by different possible offshore platforms – vessels or buoys – both in detailed simulations as well as first validation experiments. In addition to this, we have developed motion compensation algorithms that allow to correct the affected measurements and retrieve the undisturbed wind data. The motions considered and studied comprise rotations as well as translations in all six degrees of freedom.

For the evaluation of the motion-affected and corrected wind data in this paper, special attention is paid to the measurement of turbulence as well as extreme wind events. The research question to be answered is if a lidar device placed on a floating platform is capable of measuring more or less the same statistics of extreme wind events as a fixed lidar device. Quantities to be investigated are: the turbulence intensity as well as the statistics of maximum wind speed values within a 10-min period, but also wind speed increments on different time scales. At this, obviously two issues are to be discussed – the influence of the lidar measurement principle on the recording of extreme wind events, and the additional impact of the superimposed motions of the floating platform.

In order to maximize the ratio of energy capture and reduce the cost of energy, the selection of the airfoils to be used along the blade plays a crucial role. Despite the general usage of existing airfoils, more and more, families of airfoils specially tailored for specific applications are developed. The present research is focused on the design of a new family of airfoils to be used for the blade of one megawatt wind turbine working in low wind conditions. A hybrid optimization scheme has been implemented, combining together genetic and gradient based algorithms. Large part of the work is dedicated to present and discuss the requirements that needed to be satisfied in order to have a consistent family of geometries with high efficiency, high lift and good structural characteristics. For each airfoil, these characteristics are presented and compared to the ones of existing airfoils. Finally, the aerodynamic design of a new blade for low wind class turbine is illustrated and compared to a reference shape developed by using existing geometries. Due to higher lift performance, the results show a sensitive saving in chords, wetted area and so in loads in idling position.

Several methods of determining the angles of attack (AOAs) on wind turbine blades are discussed in this paper. A brief survey of the methods that have been used in the past are presented, and the advantages of each method are discussed relative to their application in the BEM theory. Data from existing as well as new full rotor CFD computations of the MEXICO rotor are used in this analysis. A more accurate estimation of the AOA is possible from 3D full rotor CFD computations, but when working with experimental data, pressure measurements and sectional forces are often the only data available. The aim of this work is to analyse the reliability of some of the simpler methods of estimating the 3D effective AOA compared some of the more rigorous CFD based methods.

In the current offshore wind turbine support structure design method, the tower and foundation, which form the support structure are designed separately by the turbine and foundation designer. This method yields a suboptimal design and it results in a heavy, overdesigned and expensive support structure. This paper presents an integrated multidisciplinary approach to design the tower and foundation simultaneously. Aerodynamics, hydrodynamics, structure and soil mechanics are the modeled disciplines to capture the full dynamic behavior of the foundation and tower under different environmental conditions. The objective function to be minimized is the mass of the support structure. The model includes various design constraints: local and global buckling, modal frequencies, and fatigue damage along different stations of the structure. To show the usefulness of the method, an existing SWT-3.6-107 offshore wind turbine where its tower and foundation are designed separately is used as a case study. The result of the integrated multidisciplinary design optimization shows 12.1% reduction in the mass of the support structure, while satisfying all the design constraints.

Measurements of mean velocity, Reynolds stresses, temperature and heat flux have been made in the wake of a model wind turbine in the EnFlo meteorology wind tunnel, for three atmospheric boundary layer states: the base-line neutral case, stable and unstable. The full-to-model scale is approximately 300:1. Primary instrumentation is two-component LDA combine with cold-wire thermometry to measure heat flux. In terms of surface conditions, the stratified cases are weak, but there is a strong 'imposed' condition in the stable case. The measurements were made between 0.5D and 10D, where D is the turbine disk diameter. In the stable case the velocity deficit decreases more slowly; more quickly in the unstable case. Heights at which quantities are maximum or minimum are greater in the unstable case and smaller in the stable case. In the stable case the wake height is suppressed but the width is increased, while in the unstable case the height is increased and the width (at hub height) reaches a maximum and then decreases. The turbulence in the wake behaves in a complex way. Further work needs to be done, to cover stronger levels of surface condition, requiring more extensive measurements to properly capture the wake development.

A series of studies are in progress investigating the effects of turbine-array-wake interactions for a range of atmospheric boundary layer states by means of the EnFlo meteorological wind tunnel. The small, three-blade model wind turbines drive 4-quadrant motor-generators. Only a single turbine in neutral flow is considered here. The motor-generator current can be measured with adequate sensitivity by means of a current sensor allowing the mean and fluctuating torque to be inferred. Spectra of torque fluctuations and streamwise velocity fluctuations ahead of the rotor, between 0.1 and 2 diameters, show that only the large-scale turbulent motions contribute significantly to the torque fluctuations. Time-lagged cross-correlation between upstream velocity and torque fluctuations are largest over the inner part of the blade. They also show the turbulence to be frozen in behaviour over the 2 diameters upstream of the turbine.

An unsteady formulation of the vortex lattice method, VLM, is presented that uses a force- free representation of the wake behind a horizontal axis wind turbine, HAWT, to calculate the aerodynamic loading on a turbine operating in the wake of an upstream rotor. A Cartesian velocity grid is superimposed over the computational domain to facilitate the representation of the atmospheric turbulence surrounding the turbine and wind shear. The wake of an upstream rotor is modelled using two methods: a mean velocity deficit with superimposed turbulence, based on experimental observations, and a purely numeric periodic boundary condition. Both methods are treated as frozen and propagated with the velocity grid.

Measurements of the mean thrust and blade root bending moment on a three bladed horizontal axis rotor modelling a 5 MW HAWT at 1:250 scale were carried out in a wind tunnel. Comparisons are made between operation in uniform flow and in the wake of a similarly loaded rotor approximately 6.5 diameters upstream. The measurements were used to validate the output from the VLM simulations, assuming a completely rigid rotor. The trends in the simulation thrust predictions are found to compare well with the uniform flow case, except at low tip speed ratios where there are losses due to stall which are yet to be included in the model. The simple wake model predicts the mean deficit, whilst the periodic boundary condition captures more of the frequency content of the loading in an upstream wake. However, all the thrust loads are over-predicted. The simulation results severely overestimate the bending moment, which needs addressing. However, the reduction in bending due to the simple wake model is found to reflect the experimental data reasonably well.

Design load simulations for wind turbines are traditionally based on the blade- element-momentum theory (BEM). The BEM approach is derived from a simplified representation of the rotor aerodynamics and several semi-empirical correction models. A more sophisticated approach to account for the complex flow phenomena on wind turbine rotors can be found in the lifting-line free vortex wake method. This approach is based on a more physics based representation, especially for global flow effects. This theory relies on empirical correction models only for the local flow effects, which are associated with the boundary layer of the rotor blades. In this paper the lifting-line free vortex wake method is compared to a state- of-the-art BEM formulation with regard to aerodynamic and aeroelastic load simulations of the 5MW UpWind reference wind turbine. Different aerodynamic load situations as well as standardised design load cases that are sensitive to the aeroelastic modelling are evaluated in detail. This benchmark makes use of the AeroModule developed by ECN, which has been coupled to the multibody simulation code SIMPACK.

CFD (Computational Fluid Dynamics) simulations are a very promising method for predicting the aerodynamic behavior of wind turbines in an inexpensive and accurate way. One of the major drawbacks of this method is the lack of validated models. As a consequence, the reliability of numerical results is often difficult to assess. The MEXICO project aimed at solving this problem by providing the project partners with high quality measurements of a 4.5 meters rotor diameter wind turbine operating under controlled conditions. The large measurement data-set allows the validation of all kind of aerodynamic models. This work summarizes our efforts for validating a CFD model based on the open source software OpenFoam. Both steady- state and time-accurate simulations have been performed with the Spalart-Allmaras turbulence model for several operating conditions. In this paper we will concentrate on axisymmetric inflow for 3 different wind speeds. The numerical results are compared with pressure distributions from several blade sections and PIV-flow data from the near wake region. In general, a reasonable agreement between measurements the and our simulations exists. Some discrepancies, which require further research, are also discussed.

Atmospheric stability strongly influences wind shear and thus has to be considered when performing load calculations for wind turbine design. Numerous methods exist however for obtaining stability in terms of the Obukhov length L as well as for correcting the logarithmic wind profile. It is therefore questioned to what extend the choice of adopted methods influences results when performing load analyses. Four methods found in literature for obtaining L, and five methods to correct the logarithmic wind profile for stability are included in the analyses (two for unstable, three for stable conditions). The four methods used to estimate stability from observations result in different PDF's of L, which in turn results in differences in estimated lifetime fatigue loads up to 81%. For unstable conditions hardly any differences are found when using either of the proposed stability correction functions, neither in wind shear nor in fatigue loads. For stable conditions however the proposed stability correction functions differ significantly, and the standard correction for stable conditions might strongly overestimate fatigue loads caused by wind shear (up to 15% differences). Due to the large differences found, it is recommended to carefully choose how to obtain stability and correct wind shear models accordingly.

Mounting a wind turbine on a floating foundation introduces more complexity to the aerodynamic loading. The floater motion contains a wide range of frequencies. To study some of the basic dynamic load effect on the blades due to these motions, a two-dimensional cascade approach, combined with a potential vortex method, is used. This is an alternative method to study the aeroelastic behavior of wind turbines that is different from the traditional blade element momentum method. The analysis tool demands little computational power relative to a full three dimensional vortex method, and can handle unsteady flows.

When using the cascade plane, a "cut" is made at a section of the wind turbine blade. The flow is viewed parallel to the blade axis at this cut. The cascade model is commonly used for analysis of turbo machineries. Due to the simplicity of the code it requires little computational resources, however it has limitations in its validity. It can only handle two-dimensional potential flow, i.e. including neither three-dimensional effects, such as the tip loss effect, nor boundary layers and stall effects are modeled. The computational tool can however be valuable in the overall analysis of floating wind turbines, and evaluation of the rotor control system.

A check of the validity of the vortex panel code using an airfoil profile is performed, comparing the variation of the lift force, to the theoretically derived Wagner function. To analyse the floating wind turbine, a floating structure with hub height 90 m is chosen. An axial motion of the rotor is considered.

For the investigation of atmospheric turbulent flows on small scales a new anemometer was developed, the so-called 2d-Atmospheric Laser Cantilever Anemometer (2d-ALCA). It performs highly resolved measurements with a spatial resolution in millimeter range and temporal resolution in kHz range, thus detecting very small turbulent structures. The anemometer is a redesign of the successfully operating 2d-LCA for laboratory application. The new device was designed to withstand hostile operating environments (rain and saline, humid air).

In February 2012, the 2d-ALCA was used for the first time in a test field. The device was mounted in about 53 m above ground level on a lattice tower near the German North Sea coast. Wind speed was measured by the 2d-ALCA at 10 kHz sampling rate and by cup anemometers at 1 Hz. The instantaneous wind speed ranged from 8 m/s to 19 m/s at an average turbulence level of about 7 %.

Wind field characteristics were analyzed based on cup anemometer as well as 2d-ALCA. The combination of both devices allowed the study of atmospheric turbulence over several magnitudes in turbulent scales.

The decay of the downstream wake of a wind turbine plays an important role in the performance of wind farms. The spread and decay of a wake depend both on wake meandering (advection of the wake as a whole) and wake diffusion (widening of the wake within its meandering frame of reference). Both of these effects depend strongly on the intensity of the ambient turbulence relative to the velocity deficit in the wake, and on the integral length scale of the turbulence relative to the wake width. Recent theory, which we review here, shows how intense large-scale turbulence can lead to a rapid x⁻² decay in the time-averaged centreline velocity deficit, as compared to a x⁻¹ decay for smaller scale turbulence, where x is distance downstream. We emphasise in this paper that common wind farm models do not predict this rapid decay. We present new experimental measurements of the velocity deficit downstream of a porous disc in relatively large-scale ambient turbulence which corroborate predictions of a x⁻² decay, and we show theoretically that the commonly used k- model does not capture this effect. We further show that a commercial CFD package, configured to match our experiments and employing the k- model, fails to predict such rapid decay. We conclude that steady simulations of wind turbine wake dynamics are insufficient for informing wind farm layout optimisation.

The performance of a model wind turbine is simulated with three different CFD methods: actuator disk, actuator line and a fully resolved rotor. The simulations are compared with each other and with measurements from a wind tunnel experiment. The actuator disk is the least accurate and most cost-efficient, and the fully resolved rotor is the most accurate and least cost-efficient. The actuator line method is believed to lie in between the two ends of the scale. The fully resolved rotor produces superior wake velocity results compared to the actuator models. On average it also produces better results for the force predictions, although the actuator line method had a slightly better match for the design tip speed. The open source CFD tool box, OpenFOAM, was used for the actuator disk and actuator line calculations, whereas the market leading commercial CFD code, ANSYS/FLUENT, was used for the fully resolved rotor approach.

Wind turbines generate electricity due to extracting energy from the wind. The rotor aerodynamics strongly depends on the flow around blade. The surface flow on the rotating blade affects the sectional performance. The wind turbine surface flow has span-wise component due to span-wise change of airfoil section, chord length, twisted angle of blade and centrifugal force on the flow. These span-wise flow changes the boundary layer on the rotating blade and the sectional performance. Hence, the thorough understanding of blade surface flow is important to improve the rotor performance. For the purpose of clarification of the flow behaviour around the rotor blade, the velocity in the boundary layer on rotating blade surface of an experimental HAWT was measured in a wind tunnel. The velocity measurement on the blade surface was carried out by a laser Doppler velocimeter (LDV). As the results of the measurement, characteristics of surface flow are clarified. In optimum tip speed operation, the surface flow on leading edge and r/R=0.3 have large span-wise velocity which reaches 20% of sectional inflow velocity. The surface flow inboard have three dimensional flow patterns. On the other hand, the flow outboard is almost two dimensional in cross sectional plane.

The purpose of this study is the Large-eddy Simulation (LES) of the turbulent wind on the complex terrain, and the first results of the simulation are described. The authors tried to apply the LES code, which was developed as an atmospheric simulator in Japan Agency for the Marine-Earth Science and Technology (JAMSTEC), to the wind prediction for the wind energy. On the wind simulation, the highest problem would be the boundary conditions, and the case in this paper was simplified one. The case study in this paper is the west wind on a complex terrain site, which is the wind from sea for the site. The steady flow was employed for the inlet condition, because the wind on the sea is the low turbulent wind, and almost all the turbulence would be generated by the roughness of the ground surface. The wall function was employed as the surface condition on the ground surface. The computational domain size was about 8 × 3 × 2.5 km³, and the minimum cell size was about 10 × 10 × 3 m³. The computational results, the vertical profile of the averaged wind speed and the turbulence intensity, agreed with the measurement by the meteorological masts. Moreover, the authors tried the analysis of the turbulence characteristics. The power spectrum density model, and the cross spectrum analyses gave the knowledge of the turbulent characteristics on the complex terrain and the hints for the domain and grid of the numerical analysis.

During last twenty years, wind turbine manufacturers took the path of building larger machines to generate more electricity. However, the bigger the size became, the more material was required to support the loads, leading to great weight increases. Larger turbines and higher hub heights also resulted in larger tower base diameters which are limited considering their logistics. In many countries, the limit for transports with special permits maximizes the diameter to 4.5 metres. Considering this fact, the wind turbine market dominated by welded steel shell towers is looking for new structural solutions for their future turbines. Although, composite materials are not used as the structural material in the towers of today's turbines, the demand for larger wind turbines forces engineers to seek for alternative material systems with high specific strength and stiffness ratios to be used in towers. Inspired by the applicability of filament winding in tower production, in the present article we investigated the effect of semi-geodesic winding on the winding angle, thickness, stiffness coefficients and vibration characteristics of filament wound composite conical shells of revolution which simulate wind turbine towers at the structural level. Present study showed that the preset friction applied during semi-geodesic winding is an important design parameter which can be controlled to obtain gradually increasing thickness from tower top to the base of the tower, and favourably alter the dynamic characteristics of the composite towers.

For wind resource assessment, the wind industry is increasingly relying on Computational Fluid Dynamics models that focus on modeling the airflow in a neutrally stratified surface layer. So far, physical processes that are specific to the atmospheric boundary layer, for example the Coriolis force, buoyancy forces and heat transport, are mostly ignored in state-of-the-art flow solvers. In order to decrease the uncertainty of wind resource assessment, the effect of thermal stratification on the atmospheric boundary layer should be included in such models. The present work focuses on non-neutral atmospheric flow over complex terrain including physical processes like stability and Coriolis force. We examine the influence of these effects on the whole atmospheric boundary layer using the DTU Wind Energy flow solver EllipSys3D. To validate the flow solver, measurements from Benakanahalli hill, a field experiment that took place in India in early 2010, are used. The experiment was specifically designed to address the combined effects of stability and Coriolis force over complex terrain, and provides a dataset to validate flow solvers. Including those effects into EllipSys3D significantly improves the predicted flow field when compared against the measurements.

This study aims to determine the best configurations of the Heat and Flux concept for more profitable and utilizable settings in a wind farm in terms of increase in the energy yield and reduction in loadings. The computations are performed with alteration of a single parameter at a time.

The reference farm for this study is EWTW, the ECN test farm in Wieringermeer, as this farm was also the reference for the validation of both the Heat and Flux concept and the software tool FarmFlow. All the studies are performed with FarmFlow developed by ECN, which computes wake deficits and turbulence intensities, resulting in the energy yield of all turbines in the farm.

Thick laminate sections can be found from the tip to the root in most common wind turbine blade designs. Obtaining accurate and reliable design data for thick laminates is subject of investigations, which include experiments on thick laminate coupons. Due to the poor thermal conductivity properties of composites and the material self-heating that occurs during the fatigue loading, high temperature gradients may appear through the laminate thickness. In the case of thick laminates in high load regimes, the core temperature might influence the mechanical properties, leading to premature failures. In the present work a method to forecast the self-heating of thick laminates in fatigue loading is presented. The mechanical loading is related with the laminate self-heating, via the cyclic strain energy and the energy loss ratio. Based on this internal volumetric heat load a thermal model is built and solved to obtain the temperature distribution in the transient state. Based on experimental measurements of the energy loss factor for 10mm thick coupons, the method is described and the resulting predictions are compared with experimental surface temperature measurements on 10 and 30mm UD thick laminate specimens.

Reliable modelling of aerodynamic induction is imperative for successful prediction of wind turbine loads and wind turbine dynamics when based on state-of- the-art aeroelastic tools. Full-scale LiDAR based wind speed measurements, with high temporal and spatial resolution, have been conducted in the rotor plane of an operating 2MW/80m wind turbine to perform detailed analysis the aerodynamic induction. The experimental setup, analyses of the spatial structure of the aerodynamic induction and subsequent comparisons with numerical predictions, using the HAWC2 aerolastic code, are presented.

Fully coupled aero-hydro-control-elastic codes are being developed to cope with the new modelling challenges presented by floating wind turbines, but there is also a place for more efficient methods of analysis. One option is linearisation and analysis in the frequency domain. For this to be an effective method, the non-linearities in the system must be well understood. The present study focusses on understanding the dynamic response of the rotor to the overall platform motion, as would arise from wave loading, by using a simple model of a floating wind turbine with a rigid tower and flexible rotor (represented by hinged rigid blades). First, an equation of motion of the blade is derived and an approximate solution for the blade response is found using the perturbation method. Secondly, the full non-linear solution is found by time- domain simulation. The response is found to be linear at lower platform pitching frequencies, becoming non-linear at higher frequencies, with the approximate solution giving good results for weakly non-linear behaviour. Higher rotor speeds have a stabilising effect on the response. In the context of typical floating turbine parameters, it is concluded that the blade flapwise response is likely to be linear.

The actuator cylinder (AC) flow model is defined as the ideal VAWT rotor. Radial directed volume forces are applied on the circular path of the VAWT rotor airfoil and constitute an energy conversion in the flow. The power coefficient for the ideal as well as the real energy conversion is defined. The describing equations for the two-dimensional AC model are presented and a solution method splitting the final solution in a linear and non-linear part is briefly described. A family of loadforms approaching the uniform loading is used to study the ideal energy conversion indicating that the maximum power coefficient for the ideal energy conversion of a VAWT could exceed the Betz limit. The real energy conversion of the 5MW DeepWind rotor is simulated with the AC flow model in combination with the blade element analysis. Aerodynamic design aspects are discussed on this basis revealing that the maximum obtainable power coefficient for a fixed pitch VAWT is constrained by the fundamental cyclic variation of inflow angle and relative velocity leading to a loading that deviates considerably from the uniform loading.

Rapid variations in the height of the recirculation zone are measured with a scanning wind lidar over a small escarpment on the Bolund Peninsula. The lidar is essentially a continuous-wave laser Doppler anemometer with the capability of rapidly changing the focus distance and the beam direction. The instrument measures the line-of-sight velocity 390 times per second and scans ten wind profiles from the ground up to seven meters per second. We observe a sharp interface between slow and fast moving fluid after the escarpment, and the interface is moving rapidly up and down. This implies that the position of the maximum velocity standard deviation is elevated a few meters above the surface. Close to the ground the mean wind is reversed relative to the general flow. The results are used to test computational fluid dynamics models for flow over terrain, and has relevance for wind energy. The preliminary comparison shows that the models are incapable of reproducing the reversed flow close to the surface, but more works needs to be done.

Current state-of-art models for floating wind turbines are built by merging separate modules addressing the four basic aspects leading to a compound hydro-servo-aero-elastic time domain solver. While current state-of-the-art models differ in many aspects, they all use the blade element momentum (BEM) aerodynamic modelling. Due to its low cost, BEM is the standard choice for design purposes. However the use of BEM entails several semi-empirical corrections and add-ons that need reconsideration and recalibration when new features appear. For floating wind turbines, the effect of the floater motions is such a new feature. In the present paper, this aspect is investigated by comparing BEM based results against 3D free-wake simulations. Deterministic as well as stochastic simulations are presented in pure aerodynamic and full aeroelastic context. It is confirmed that asymmetric inflow originating from yaw misalignment and shear give significant differences reflected on mean values and amplitudes.

Stall Cells (SCs) are large scale three-dimensional structures of separated flow that have been observed on the suction side of airfoils designed for or used on wind turbine blades. SCs are unstable in nature but can be stabilised by means of a localized disturbance; here in the form of a zigzag tape covering 10% of the wing span. Based on extensive tuft flow visualisations, the resulting flow was found macroscopically similar to the undisturbed flow. Next a combined investigation was carried out including pressure recordings, Stereo-PIV measurements and CFD simulations. The investigation parameters were the aspect ratio, the angle of attack and the Re number. Tuft and pressure data were found in good agreement. The 3D CFD simulations reproduced the structure of the SCs in qualitative agreement with the experimental data but had a delay of ~3deg in capturing the first appearance of a SC. The error in Cl max prediction was 7% compared to 19% for the 2D cases. Tests show that SCs grow with Re number and angle of attack. Also analysis of the time averaged computational results indicated the presence of three types of vortices: (a) the trailing edge line vortex (TELV) in the wake, (b) the separation line vortex (SLV) over the wing and (c) the SC vortices. The TELV and SLV run parallel to the trailing edge and are of opposite sign, while the SC vortices start normal to the wing suction surface, then bend towards the SC centre and later extend downstream, with their vorticity parallel to the free stream.

Design verification of wind turbines is performed by simulation of design load cases (DLC) defined in the IEC 61400-1 and -3 standards or equivalent guidelines. Due to the resulting large number of necessary load simulations, here a method is presented to reduce the computational effort for DLC simulations significantly by introducing a reduced nonlinear model and simplified hydro- and aerodynamics. The advantage of the formulation is that the nonlinear ODE system only contains basic mathematic operations and no iterations or internal loops which makes it very computationally efficient. Global turbine extreme and fatigue loads such as rotor thrust, tower base bending moment and mooring line tension, as well as platform motions are outputs of the model. They can be used to identify critical and less critical load situations to be then analysed with a higher fidelity tool and so speed up the design process. Results from these reduced model DLC simulations are presented and compared to higher fidelity models. Results in frequency and time domain as well as extreme and fatigue load predictions demonstrate that good agreement between the reduced and advanced model is achieved, allowing to efficiently exclude less critical DLC simulations, and to identify the most critical subset of cases for a given design. Additionally, the model is applicable for brute force optimization of floater control system parameters.

The aim of this work is to assess the prediction capabilities of the turbulence models and the transition model kkl-ω available in OpenFOAM and to achieve a database of airfoil aerodynamical characteristics. The airfoils chosen for the simulations are FX 79-W- 15A and NACA 63-430, which are widely used in wind turbines. The numerically obtained lift and drag coefficients are compared with available experimental results. A quantitative and qualitative study is conducted to determine the influence of meshing strategies, computational time step together with interpolation and temporal schemes. Two Reynolds Averaged Navier- Stokes models (RANS models) are used, which are the k-ω SST model by Menter and the kkl-ω model (which involves transition modeling) by Walters and Davor.

This article presents detached eddy simulation (DES) results of a 5MW wind turbine in an unsteady atmospheric boundary layer. The evaluation performed in this article focuses on turbine blade loads as well as on the influence of atmospheric turbulence and tower on blade loads. Therefore, the turbulence transport of the atmospheric boundary layer to the turbine position is analyzed. To determine the influence of atmospheric turbulence on wind turbines the blade load spectrum is evaluated and compared to wind turbine simulation results with uniform inflow. Moreover, the influences of different frequency regimes and the tower on the blade loads are discussed. Finally, the normal force coefficient spectrum is analyzed at three different radial positions and the influence of tower and atmospheric turbulence is shown.

Since the 1970s several research activities had been carried out on developing aerodynamic models for Vertical Axis Wind Turbines (VAWTs). In order to design large VAWTs of MW scale, more accurate aerodynamic calculation is required to predict their aero-elastic behaviours. In this paper, a 3D free wake vortex lattice model for VAWTs is developed, verified and validated. Comparisons to the experimental results show that the 3D free wake vortex lattice model developed is capable of making an accurate prediction of the general performance and the instantaneous aerodynamic forces on the blades. The comparison between momentum method and the vortex lattice model shows that free wake vortex models are needed for detailed loads calculation and for calculating highly loaded rotors.

There is renewed interest in aerodynamics research of VAWT rotors. Lift type, Darrieus designs sometimes use flexed blades to have an 'egg-beater shape' with an optimum Troposkien geometry to minimize the structural stress on the blades. While straight bladed VAWTs have been investigated in depth through both measurements and numerical modelling, the aerodynamics of flexed blades has not been researched with the same level of detail.

Two major effects may have a substantial impact on blade performance. First, flexing at the equator causes relatively strong trailing vorticity to be released. Secondly, the blade performance at each station along the blade is influenced by self-induced velocities due to bound vorticity. The latter is not present in a straight bladed configuration. The aim of this research is to investigate these effects in relation to an innovative 4kW wind turbine concept being developed in collaboration with industry known as a self-adjusting VAWT (or SATVAWT). The approach used in this study is based on experimental and numerical work. A lifting line free-wake vortex model was developed. Wind tunnel power and hot-wire velocity measurements were performed on a scaled down, 60cm high, three bladed model in a closed wind tunnel.

Results show a substantial axial wake induction at the equator resulting in a lower power generation at this position. This induction increases with increasing degree of flexure. The self-induced velocities caused by blade bound vorticity at a particular station was found to be relatively small.

The tip vortex of a wind turbine rotor blade originates as a result of a complex distribution of vorticity along the blade tip thickness. While the tip vortex evolution was extensively studied previously in other work, the mechanism of the initiation of the tip vorticity in a 3D rotating environment is still somewhat obscured due to lack of detailed experimental evidence. This paper therefore aims at providing an understanding of how tip vorticity is formed at the wind turbine blade tip and what happens just behind the tip trailing edge. Stereo Particle Image Velocimetry (SPIV) is used to measure the flow field at the tip of a 2m diameter, two- bladed rotor at the TU Delft Open Jet Facility (OJF). The rotor has a rectangular blade tip. Spanwise measurements were performed for both axial and yawed flow conditions with a very small azimuthal increment. A 3D, unsteady, potential flow panel method is also used for the purpose of better understanding the tip bound vorticity. A validation study is carried out with positive results. This paper is focused on axial flow results.

A complex distribution of vorticity is found along the blade tip thickness. Just after release, the tip vortex becomes almost immediately round and well defined. Observations from the MEXICO rotor are confirmed again by a slight inboard convection of the tip vortex. This is explained by means of the effect of chordwise vorticity at the tip from the numerical solutions.

The results presented in this work suggest that a more physical interpretation of the tip loss effect is required. Currently, inclusion of tip effects are based primarily on either wake induced effects or on an empirical 3D correction for airfoil data. This research should stimulate a more rigorous approach, where the effects of the blade tip chordwise vorticity are implemented in tip correction models.

Support structures for offshore wind turbines are contributing a large part to the total project cost, and a cost saving of a few percent would have considerable impact. At present support structures are designed with simplified methods, e.g., spreadsheet analysis, before more detailed load calculations are performed. Due to the large number of loadcases only a few semimanual design iterations are typically executed. Computer-assisted optimization algorithms could help to further explore design limits and avoid unnecessary conservatism.

In this study the simultaneous perturbation stochastic approximation method developed by Spall in the 1990s was assessed with respect to its suitability for support structure optimization. The method depends on a few parameters and an objective function that need to be chosen carefully. In each iteration the structure is evaluated by time-domain analyses, and joint fatigue lifetimes and ultimate strength utilization are computed from stress concentration factors. A pseudo-gradient is determined from only two analysis runs and the design is adjusted in the direction that improves it the most.

The algorithm is able to generate considerably improved designs, compared to other methods, in a few hundred iterations, which is demonstrated for the NOWITECH 10 MW reference turbine.

The parameters influencing the long term extreme operating design loads are identified through the implementation of a Design of Experiment (DOE) method. A function between the identified critical factors and the ultimate out-of-plane loads on the blade is determined. Variations in the initial blade azimuth location are shown to affect the extreme blade load magnitude during operation in normal turbulence wind input. The simultaneously controlled operation of generator torque variation and pitch variation at low blade pitch angles is detected to be responsible for very high loads acting on the blades. Through gain scheduling of the controller (modifications of the proportional Kp and the integral K gains) the extreme loads are mitigated, ensuring minimum instantaneous variations in the power production for operation above rated wind speed. The response of the blade load is examined for different values of the integral gain as resulting in rotor speed error and the rate of change of rotor speed. Based on the results a new load case for the simulation of extreme loads during normal operation is also presented.

The wake behind a three-bladed Glauert model rotor in a water channel was investigated. Planar particle image velocimetry was used to measure the velocity fields on the wake centre-line, with snapshots phase-locked to blade position of the rotor. Phase- locked averages of the velocity and vorticity fields are shown, with tip vortex interaction and entanglement of the helical filaments elucidated. Proper orthogonal decomposition and topology-based vortex identification are used to filter the PIV images for coherent structures and locate vortex cores. Application of these methods to the instantaneous data reveals unsteady behaviour of the helical filaments that is statistically quantifiable.

As wind turbine blades become larger there is a tendency for the blade torsional stiffness to reduce, producing the possibility of dynamic instability at moderate windspeeds.

While linearised methods can assess the envelope of allowable blade properties for avoiding classical flutter with attached flow aerodynamics, wind turbine aerofoils can experience stalled flow. Therefore, it is necessary to explore the possible effects of stall-flutter on blade stability. This paper aims to address methods for judging the stability of blade designs during both attached flow and stalled flow behaviour.

This paper covers the following areas:

i) Attached flow model

A Beddoes-Leishman indicial model is presented and the choice of coefficients is explained in the context of Theodorsen's theory for flat-plate aerofoils and experimental results by Beddoes and Leishman. Special attention is given to the differing dynamic behaviour of the pitching moment due to flapping motion, pitching motion and dynamically varying inflow.

(ii) Classical flutter analysis

The time domain attached flow model is verified against a linear flutter analysis by comparing time domain results for a 3D model of a representative multi-megawatt turbine blade, varying the position of the centre of mass along the chord. The results show agreement to within 6% for a range of flutter onset speeds.

(iii) Dynamic stall model

On entering the stalled region, damping of torsional motion of an aerofoil section can become negative. A dynamic stall model which encompasses the effects of trailing edge separation and leading edge vortex detachment is presented and validated against published experimental data.

(iv) Stall flutter

The resulting time domain model is used in simulations validating the prediction of reduced flutter onset for stalled aerofoils. Representative stalled conditions for a multi-megawatt wind turbine blade are investigated to assess the possible reduction in flutter speed. A maximum reduction of 17% is observed for the particular blade investigated.

The effects of a rough surface boundary layer on the outputs of a wind turbine model were investigated experimentally in a wind tunnel. The very rough surface consisted of cylindrical pins, in order to model a forest canopy. The hub height of the turbine model was varied in order to see the effect of the presence of the model forest in the power and thrust coefficients. A small effect of the hub height was observed in the averaged power coefficient, where the turbine produced less for the lowest hub height. The difference was however reduced when scaling the power output with the available power in the wind instead of using the velocity at hub height. Consistent trends were present in the standard deviation of the thrust coefficient and the rotational speed, which both increased by decreasing the hub height. This underlines the fact that not only the rotor but also the tower and the bearings of a wind turbine must withstand to increased loads when operating close to a canopy.

The flow behind the model of wind turbine rotor is investigated experimentally in a water flume using Particle Image Velocimetry. The study carried out involves rotors of three bladed wind turbine designed using Glauert's optimization. The transitional regime, generally characterized as in between the regime governed by stable organized vortical structures and the turbulent wake, develops from disturbances of the tip and root vorticies through vortex paring and further complex behaviour towards the fully turbulent wake. Our PIV measurements pay special attention to the onset of the instabilities. The near wake characteristics (development of expansion, tip vortex position, deficit velocity and rotation in the wake) have been measured for different tip speed ratio to compare with main assumptions and conclusions of various rotor theories.

An Actuator Disk (AD) model is implemented in the CFD platform OpenFOAM^® with the purpose of studying the characteristics of the turbulent flow in the wake of the rotor of a horizontal-axis wind turbine. This AD model is based on the blade-element theory and it employs airfoil data to calculate the distribution of forces over the disk of a conceptual 5 MW offshore wind turbine. A uniform, non-turbulent flow is used as inflow so the turbulence is only produced in the wake of the AD. Computations are performed using Large-Eddy Simulations (LES) to capture the unsteady fluctuations in the flow. Additionally, a classic Smagorinsky Sub-Grid Scale (SGS) technique is employed to model the unfiltered motions. This new AD implementation makes use of a control system to adjust the rotational velocity of the rotor (below rated power) to the local conditions of the wind flow. The preliminary results show that the wake characteristics are influenced by the force distribution on the disk when compared to the wake produced by a uniformly loaded AD. Also, we observe that the simulated rotor reacts correctly to the introduction of the control system, although operating below the optimal power.

The increasing size of Horizontal Axis Wind Turbines and the trend to install wind farms further offshore demand more robust design options. If the pitch system could be eliminated, the availability of Horizontal Axis Wind Turbines should increase. This research investigates the use of active stall control to regulate power production in replacement of the pitch system. A feasibility study is conducted using a blade element momentum code and taking the National Renewable Energy Laboratory 5 MW turbine as baseline case. Considering half of the blade span is equipped with actuators, the required change in the lift coefficient to regulate power was estimated in ΔC_l = 0.7. Three actuation technologies are investigated, namely Boundary Layer Transpiration, Trailing Edge Jets and Dielectric Barrier Discharge actuators. Results indicate the authority of the actuators considered is not sufficient to regulate power, since the change in the lift coefficient is not large enough. Active stall control of Horizontal Axis Wind Turbines appears feasible only if the rotor is re-designed from the start to incorporate active-stall devices.

A near wake model, originally proposed by Beddoes, is further developed. The purpose of the model is to account for the radially dependent time constants of the fast aerodynamic response and to provide a tip loss correction. It is based on lifting line theory and models the downwash due to roughly the first 90 degrees of rotation. This restriction of the model to the near wake allows for using a computationally efficient indicial function algorithm. The aim of this study is to improve the accuracy of the downwash close to the root and tip of the blade and to decrease the sensitivity of the model to temporal discretization, both regarding numerical stability and quality of the results. The modified near wake model is coupled to an aerodynamics model, which consists of a blade element momentum model with dynamic inflow for the far wake and a 2D shed vorticity model that simulates the unsteady buildup of both lift and circulation in the attached flow region. The near wake model is validated against the test case of a finite wing with constant elliptical bound circulation. An unsteady simulation of the NREL 5 MW rotor shows the functionality of the coupled model.

A simulation study on the wind field resolution in computer load simulations has been conducted, both in transversal/vertical and longitudinal direction, to determine the effect on blade fatigue loading. Increasing the transversal/vertical resolution decreased the loading significantly, while only small changes to the load, at very low frequencies were found for increased longitudinal resolution. Next the influence of the tower shadow for a downwind mounted rotor was investigated, with respect to blade fatigue loading. The influence of different components to the total tower shadow effect was studied, both for a monopile and a truss tower, latter at inclination 0 and 22.5 degrees with respect to the incoming wind direction. Four components were considered, both individually and in combinations: mean wind speed, mean velocity deficit, unsteady motions from vortex shedding, and turbulence. The mean velocity deficit and turbulence were the main contributors to blade fatigue loading, and the unsteady motions can be neglected for the truss tower. For the monopile, neglecting the unsteady motions resulted in an underestimation of fatigue loading in the order of 3 percent.

The following study proposes a two-dimensional large-scale particle tracking velocimetry (LS-PTV) system to characterize coherent wind structures. Seven minutes of LS-PTV data is collected via an apparatus that seeds fog-filled soap bubbles into the wind at a height of 6m from the ground. The LS-PTV data is compared to 20 minutes of data collected concurrently from a wind mast at the same site. The LS-PTV system recorded a mean streamwise velocity of 1.35m/s with a standard deviation of 0.23m/s at a mean height of 2.50m with a standard deviation of 0.7m, which agrees well with the velocity profile measured by the wind mast. Furthermore, the Reynolds stresses measured by the LS-PTV system are found to compare to those measured by the wind mast and by Klebanoff [1] for a canonical turbulent boundary layer. The current study assumes that the centre-of-curvature trajectories of the particle pathlines are representative of the trajectories followed by the spanwise vortices. As a proof-of-principle study, this work has been successful in accurately describing the vortex distribution very near to the ground. However, the trajectories followed by the centres-of- curvat.ure belonging to pathlines concurrently passing through the field-of-view were sporadic and uncorrelated.

Many Australian wind farms are located near escarpments and cliffs where flow separation occurs. An absence of literature addressing the effect of wind direction over cliffs have motivated surface shear stress visualisations on forward facing steps at yaw angles between 0° and 50°. These visualisations have been conducted in the Monash University 450 kW wind tunnel. Mean reattachment lengths were measured and shown to vary as a function of the boundary layer thickness to step height ratio and the yaw angle. Vortices shed off the crest of the step induced surface shear stresses on the top surface of the step. The orientation of these shear stresses varied linearly with the yaw angle. Three-dimensional structures of different forms were also observed. At zero yaw angle the flow converged at points along the crest. At high yaw angles distinct sections of misaligned flow were observed downstream of the reattachment line, indicating a spatial periodicity in shedding.

Large eddy simulation (LES) of an infinitely long wind farm in a fully developed flow is carried out based on solution of the incompressible Navier-Stokes equations. The wind turbines are modeled as equivalent rotating actuator disks by applying aerodynamic loads on the flow field using tabulated aerodynamic lift and drag coefficients to save computational time. As a substitute to standard wall modeling LES, a ''prescribed mean shear" profile (hereafter called PMS) approach has been implemented and analysed for generating the desired turbulent shear flow. It is applied on Neutral, Stable and Convective atmospheric boundary layers in presence of the -actuator disc represented- wind turbines and qualitatively meaningful results of mean and fluctuating velocity field is obtained. The effect of four different sub-grid scale (SGS) models on the flow structure is investigated and it is seen that subgrid scale modeling (in particular, the Mix-O and Smagorinsky models) improves the accuracy of the simulations. An optimal grid resolution is also proposed for this kind of simulation.

In recent years many advanced load simulation tools, allowing an aero-servo-hydroelastic analyses of an entire offshore wind turbine, have been developed and verified. Nowadays, even an offshore wind turbine with a complex support structure such as a jacket can be analysed. However, the computational effort rises significantly with an increasing level of details. This counts especially for offshore wind turbines with lattice support structures, since those models do naturally have a higher number of nodes and elements than simpler monopile structures. During the design process multiple load simulations are demanded to obtain an optimal solution. In the view of pre-design tasks it is crucial to apply load simulations which keep the simulation quality and the computational effort in balance. The paper will introduce a reference wind turbine model consisting of the REpower5M wind turbine and a jacket support structure with a high level of detail. In total twelve variations of this reference model are derived and presented. Main focus is to simplify the models of the support structure and the foundation. The reference model and the simplified models are simulated with the coupled simulation tool Flex5-Poseidon and analysed regarding frequencies, fatigue loads, and ultimate loads. A model has been found which reaches an adequate increase of simulation speed while holding the results in an acceptable range compared to the reference results.

The paper presents the final results from the first phase of IEA Task 29 'Mexnext'. Mexnext was a joint project in which 20 parties from 11 different countries cooperated. The main aim of Mexnext was to analyse the wind tunnel measurements which have been taken in the EU project 'MEXICO'. In the MEXICO project 10 institutes from 6 countries cooperated in doing experiments on an instrumented, 3 bladed wind turbine of 4.5 m diameter placed in the 9.5 by 9.5 m² open section of the Large Low-speed Facility (LLF) of DNW in the Netherlands.

Pressure distributions on the blades were obtained from 148 Kulite pressure sensors, distributed over 5 sections at 25, 35, 60, 82 and 92 % radial position respectively. Blade loads were monitored through two strain-gauge bridges at each blade root. Most interesting however are the extensive PIV flow field measurements, which have been taken simultaneously with the pressure and load measurements. As a result of the international collaboration within this task a very thorough analysis of the data could be carried out and a large number of codes were validated not only in terms of loads but also in terms of underlying flow field.

The paper will present several results from Mexnext-I, i.e. validation results and conclusion on modelling deficiencies and directions for model improvement. The future plans of the Mexnext consortium are also briefly discussed. Amongst these are Mexnext-II, a project in which also aerodynamic measurements other than MEXICO are included, and 'New MEXICO' in which additional measurement on the MEXICO model are performed.

This work presents the results from a field test of LIDAR assisted collective pitch control using a scanning LIDAR device installed on the nacelle of a mid-scale research turbine. A nonlinear feedforward controller is extended by an adaptive filter to remove all uncorrelated frequencies of the wind speed measurement to avoid unnecessary control action. Positive effects on the rotor speed regulation as well as on tower, blade and shaft loads have been observed in the case that the previous measured correlation and timing between the wind preview and the turbine reaction are accomplish. The feedforward controller had negative impact, when the LIDAR measurement was disturbed by obstacles in front of the turbine. This work proves, that LIDAR is valuable tool for wind turbine control not only in simulations but also under real conditions. Furthermore, the paper shows that further understanding of the relationship between the wind measurement and the turbine reaction is crucial to improve LIDAR assisted control of wind turbines.

Wind measurements of a 2D Multi-Lidar and a mast mounted cup anemometer are compared in this study. Average wind speed and direction as well as the turbulence intensity of the wind speed are considered. Data analysis is mainly performed using standard regression analysis on 10 minute average data and the calculation of the power spectral density. The results show a good agreement regarding wind speed and direction and the turbulence intensity of the horizontal wind.

In the present study the aerodynamic boundary layer at a rotor blade is investigated while the turbine is working under real operating conditions in the atmosphere. Owing to the complexity of the experimental set-up, up to now most research on transition is conducted in wind tunnels and field measurements are rare. Hence important effects such as the unsteady behavior of the inflow is not taken into account. For the current measurements the blade is equipped with a hot film at the most interesting part of the upper side midspan of the blade in order to detect non-laminar structures in the boundary layer. Furthermore, 34 pressure tubes are installed along the chord length in order to gain information about the flow field. A preliminary analysis of the hot-film measurements combined with a CFD calculation and a stability analysis based on the e^N method leads to two results. Firstly it is possible to determine the state of the boundary layer (laminar or turbulent) and secondly we propose to discuss our findings in case of medium rotational speed within so called Tollmien-Schlichting scenario.

The CQU-DTU-LN1 series of airfoils were designed with an objective of high lift and low noise emission. In the design process, the aerodynamic performance is obtained using XFOIL while noise emission is obtained with the BPM model. In this paper we present some validations of the designed CQU-DTU-LN118 airfoil by using wind tunnel measurements in the acoustic wind tunnel located at Virginia Tech and numerical computations with the inhouse Q³uic and EllipSys 2D/3D codes. To show the superiority of the new airfoils, comparisons with a NACA64618 airfoil are made. For the aerodynamic features, the designed Cl and Cl/Cd agrees well with the experiment and are in general higher than those of the NACA airfoil. For the acoustic features, the noise emission of the LN118 airfoil is compared with the acoustic measurements and that of the NACA airfoil. Comparisons show that the BPM model can predict correctly the noise changes.

Tip loss correction is known to play an important role for engineering prediction of wind turbine performance. There are two different types of tip loss corrections: tip corrections on momentum theory and tip corrections on airfoil data. In this paper, we study the latter using detailed CFD computations for wind turbines with sharp tip. Using the technique of determination of angle of attack and the CFD results for a NordTank 500 kW rotor, airfoil data are extracted and a new tip loss function on airfoil data is derived. To validate, BEM computations with the new tip loss function are carried out and compared with CFD results for the NordTank 500 kW turbine and the NREL 5 MW turbine. Comparisons show that BEM with the new tip loss function can predict correctly the loading near the blade tip.

Neural network algorithms have shown the capability to infer the actual wind turbine loading from standard signals commonly used for operational control purposes. Fatigue load monitoring done with this readily available data, can offer a robust and cost effective alternative to conventional maintenance-intensive mechanical stress measurement devices. The concept needs to be adopted to offshore wind turbines, where the exposure to the harsh environment with rather difficult accessibility makes the use particularly attractive. At such a site the impact of hydro-dynamically dominated loads might result in poor fatigue estimates, which is due to the lack of information on the surrounding sea state. In order to avoid the need of measuring-buoys, this work studies the employment of additional accelerometers mounted directly at the structure. Various potential placements and three sub-structure types are considered to account for the characteristic structural response caused by wave induced loading. The feasibility is demonstrated on generic data, gained from simulations. Recommended practices are deduced and applied to data from the AREVA M5000 turbine at "alpha ventus".

In the present article we describe CFD simulations of the well known NREL Phase-VI rotor in axial flow conditions using a newly developed technique of combining turbulence modeling by the Delayed Detached Eddy Simulation (DDES) technique with laminar/turbulent transition modeling by a correlation based method. We demonstrate how the power production around the onset of stall is very dependent on the turbulence intensity in the inflow. Additionally, we compare with measurements and illustrate how the unsteady loads from the DDES simulations can provide valuable insight in the transient behavior of the rotor loads.

We carry out large-eddy simulation of turbulent flow past a complete hydrokinetic turbine mounted on the bed of a straight rectangular open channel. The complex turbine geometry, including the rotor and all stationary components, is handled by employing the curvilinear immersed boundary (CURVIB) method [1], and velocity boundary conditions near all solid surfaces are reconstructed using a wall model based on solving the simplified boundary layer equations [2]. In this study we attempt to directly resolve flow-blade interactions without introducing turbine parameterization methods. The computed wake profiles of velocities and turbulent stresses agree well with the experimentally measured values.

A wind tunnel experiment was performed to study turbulence processes within a model wind turbine array of 3 by 8 model wind turbines of alternating sizes placed aligned with the mean flow. The model wind farm was placed in a boundary layer developed over both smooth and rough surfaces under neutrally stratified conditions. Turbulence statistics, TKE budget terms, and the spectral structure of the turbulence generated within and above the wind farm reveal relevant information about the processes modulating the turbulent energy transfer from the boundary layer to the turbines. The results of the experiment suggest that heterogeneity in turbine size within a wind farm introduce complex flow interactions not seen in a homogeneous farm, and may have positive effects on turbulent loading on the turbines and turbulent exchange with the atmosphere. In general, large scale motions are heavily dampened behind the first row of turbines but a portion of such structures are generated far inside the wind farm, and the scale of the most energetic eddy motions was relatively consistent at different elevations. Overall, the experiment revealed the possibility that heterogeneity of wind turbine size within wind farms have the potential to change the overall potential to harvest energy from the wind, and alter the economics of a project.

Fine tuning of controllers for pitch-torque regulated wind turbines is an opportunity to improve the wind turbine performances and reduce the cost of energy without applying any changes to the design. For this purpose, a method for automatically tune a classical controller based on numerical optimization is developed and tested. To have a better understanding of the problem a parametric analysis of the wind turbine performances due to changes in the controller parameters is first performed. Thereafter results obtained with the automatic tuning show that is possible to identify a finer controller tuning that improves the wind turbine performances. For the case study selected in this work, a 2% cost function reduction is achieved with seven iterations.

The wake of the 2MW NM80 wind turbine subject to non-uniform and laminar inflow conditions is simulated using CFD with a fully resolved rotor geometry, an actuator line method and actuator disc method, respectively and in all simulations the wake properties are compared. Based on the comparison the strengths and limitations of the models are pointed out.

Most state of the art rotor design methods are based on the actuator disc theory developed about one century ago. The actuator disc is an axisymmetric permeable surface carrying a load that represents the load on a real rotor with a finite number of blades N. However, the mathematics of the transition from a real rotor load to an axisymmetrically loaded disc is not yet presented in literature. By formulating an actuator disc equation of motion in which the Bernoulli constant H is expressed in kinematical terms, a comparison of the power conversion and load on the disc and rotor is possible. For both the converted power is expressed as a change of angular momentum times rotational speed. The limits for N → ∞ while the chord c → 0, the rotational speed Ω → ∞, the load F becoming uniform by ∂F/∂r → 0 and the thickness → 0 confirm that the classical disc represents the rotor with an infinite number of blades. Furthermore, the expressions for the blade load are compared to the expressions in current design and analysis tools. The latter do not include the load on chord-wise vorticity. Including this is expected to give a better modelling of the tip and root flow.

Stereoscopic Particle Image Velocimetry measurements have been conducted in cross-planes behind three different geometries of Vortex Generators (VGs) in a high Reynolds number boundary layer. The VGs have been mounted in a cascade producing counter-rotating vortices and the downstream flow development was examined. Three VG geometries were investigated: rectangular, triangular and cambered. The various VG geometries tested are seen to produce different impacts on the boundary layer flow. Helical symmetry of the generated vortices is confirmed for all investigated VG geometries in this high Reynolds number boundary layer. From the parameters resulting from this analysis, it is observed at the most upstream measurement position that the rectangular and triangular VGs produce vortices of similar size, strength and velocity induction whilst the cambered VGs produce smaller and weaker vortices. Studying the downstream development in the ensemble and spanwise averaged measurements, it is observed that the impact from the rectangular and triangular VGs differs. For the rectangular VGs, self-similarity in the streamwise component was confirmed.

This research paper presents preliminary results on a behavioural study of a free yawing downwind wind turbine. A series of wind tunnel tests was performed at the TU Delft Open Jet Facility with a three bladed downwind wind turbine and a rotor radius of 0.8 meters. The setup includes an off the shelf three bladed hub, nacelle and generator on which relatively flexible blades are mounted. The tower support structure has free yawing capabilities provided at the base. A short overview on the technical details of the experiment is given as well as a brief summary of the design process. The discussed test cases show that the turbine is stable while operating in free yawing conditions. Further, the effect of the tower shadow passage on the blade flapwise strain measurement is evaluated. Finally, data from the experiment is compared with preliminary simulations using DTU Wind Energy's aeroelastic simulation program HAWC2.

The aerodynamic performance of flat-back and elliptically shaped airfoils is analyzed on the basis of CFD simulations. Incompressible and low-Mach preconditioned compressible unsteady simulations have been carried out using the k-w SST and the Spalart Allmaras turbulence models. Time averaged lift and drag coefficients are compared to wind tunnel data for the FB 3500-1750 flat back airfoil while amplitudes and frequencies are also recorded. Prior to separation averaged lift is well predicted while drag is overestimated keeping however the trend in the tests. The CFD models considered, predict separation with a 5° delay which is reflected on the load results. Similar results are provided for a modified NACA0035 with a rounded (elliptically shaped) trailing edge. Finally as regards the dynamic characteristics in the load signals, there is fair agreement in terms of Str number but significant differences in terms of lift and drag amplitudes.

In ECN's scaled wind farm the wake evolution is studied in two different situations. A single wake is studied at two different locations downstream of a turbine and a single wake is studied in conjunction with a triple wake. Here, the wake is characterized by the relative wind speed, the turbulence intensity, the vertical wind speed and the turbulence (an)isotropy. Per situation all wake measurements are taken simultaneously together with the inflow conditions.

Wind and power deficit in the wake are assessed for the offshore wind farm Alpha Ventus. Operational data are evaluated for the power deficit in the wake of a single wind turbine and in a row of wind turbines. The wake of a single wind turbine is described by the maximum power deficit and expansion width of the wake. The impact of atmospheric stability in respect to vertical wind shear and turbulence intensity is assessed showing that wake effects are more pronounced under stable conditions.

The effect of Constant Life Diagram (CLD) formulation on the fatigue life prediction under variable amplitude (VA) loading was investigated based on variable amplitude tests using three different load spectra representative for wind turbine loading. Next to the Wisper and WisperX spectra, the recently developed NewWisper2 spectrum was used. Based on these variable amplitude fatigue results the prediction accuracy of 4 CLD formulations is investigated. In the study a piecewise linear CLD based on the S-N curves for 9 load ratios compares favourably in terms of prediction accuracy and conservativeness. For the specific laminate used in this study Boerstra's Multislope model provides a good alternative at reduced test effort.

In this study, high-resolution large-eddy simulations of two German offshore wind farms are performed with the LES model PALM in which the turbines are parameterized with an actuator disk approach. The simulation of a single wind farm is realized by a stationary model domain with a turbulent inflow. Different atmospheric stratifications from slightly stable to unstable have been simulated and their effect on the wake characteristics is investigated. The results show a clear development of the flow within the wind farms with large differences between single, double and triple wakes. The wake deficit is increasing up to the second wake for the neutral and stable boundary layer and up to the third wake in the convective case before reaching a nearly constant value. This is explained with the turbulence intensity being much higher behind the second and subsequent turbines compared to the wake of the first turbine. With increasing atmospheric stability the wake deficits are shown to be stronger due to reduced turbulence.

The layouts of turbines affect the turbine wake interactions and thus the wind farm performance. The wake interactions in infinite staggered wind-turbine arrays are investigated and compared with infinite aligned turbine arrays in this paper. From the numerical results we identify three types of wake behaviours, which are significantly different from wakes in aligned wind-turbine arrays. For the first type, each turbine wake interferes with the pair of staggered downstream turbine wakes and the aligned downstream turbine. For the second type, each turbine wake interacts with the first two downstream turbine wakes but does not show significant interference with the second aligned downstream turbine. For the third type, each turbine wake recovers immediately after passing through the gap of the first two downstream turbines and has little interaction with the second downstream turbine wakes The extracted power density and power efficiency are also studied and compared with aligned wind-turbine arrays.

Accurate load simulations are necessary in order to design cost-efficient support structures for offshore wind turbines. Due to software limitations and confidentiality issues, support structures are often designed with sequential analyses, where simplified wind turbine and support structure models replace more detailed models. The differences with an integrated analysis are studied here for a commercial OWEC Quattropod. Integrated analysis seems to generally predict less damage than sequential analysis, decreasing by 30-70 percent in two power production cases with small waves.

Additionally it was found that using a different realization of the wave forces for the retrieval run in sequential analysis leads to an increase of predicted damage, which can be explained as the effect of applying two independent wave force series at the same time.

The midsection of the detailed support structure model used shell elements. Additional analyses for a model with an equivalent beam model of the midsection showed only small differences, mostly overpredicting damage by a few percent. Such models can therefore be used for relatively accurate analysis, if carefully calibrated.

In the present study, a simple inviscid vortex ring (VR) modelling approach is used to represent the developing rotor wake. This allows a straightforward investigation and comparison of the impact of uniform, yawed and sheared flow conditions on the development of the rotor wake, with the additional possibility of including ground effect. The effect of instabilities on the development of the wake is manually introduced in the form of perturbations of strength, ring position and size. The phenomenon of vortex filament interaction or leapfrogging, could play a role in the observation of unsteady phenomena and is therefore also addressed. Such a study is hence performed in light of recent conflicting views on the causes of wake meandering: is the observed dynamic wake behaviour a result of large scale turbulent forcing or do more subtle and intrinsic wake instabilities play a role? This study concludes that the presence of the ground and external perturbations, most notably changes in the wake pitch and the rotor thrust coefficient, can significantly affect the steady development of the wake. The mutual vortex pairing instability, whilst displaying interesting periodic behaviour, does not correlate with periodic wake behaviour reported by Medici et al. [1]. However, in the absence of unsteady inflow, it is shown that the wake of a Horizontal Axis Wind Turbine (HAWT) is certainly prone to displaying unstable, dynamic behaviour caused by these additional factors.

Turbine mounted scanning lidar systems of focussed continuous-wave type are taken into consideration to sense approaching wind fields. The quality of wind information depends on the lidar technology itself but also substantially on the scanning technique and reconstruction algorithm. In this paper a five-parameter wind field model comprising mean wind speed, vertical and horizontal linear shear and homogeneous direction angles is introduced. A corresponding parameter estimation method is developed based on the assumption of upwind lidar measurements scanned over spherical segments. As a main advantage of this method all relevant parameters, in terms of wind turbine control, can be provided. Moreover, the ability to distinguish between shear and skew potentially increases the quality of the resulting feedforward pitch angles when compared to three-parameter methods. It is shown that minimal three measurements, each in turn from two independent directions are necessary for the application of the algorithm, whereas simpler measurements, each taken from only one direction, are not sufficient.