0
Research Papers: Ocean Renewable Energy

# Characterization of a Wind Generation System for Use in Offshore Wind Turbine DevelopmentOPEN ACCESS

[+] Author and Article Information
Raul Urbina, James M. Newton, Matthew P. Cameron, Andrew J. Goupee, Krish P. Thiagarajan

Department of Mechanical Engineering,
University of Maine,
5711 Boardman Hall,
Orono, ME 04469

Richard W. Kimball

301 Dismukes Hall,
Castine, ME 04420

Contributed by the Ocean, Offshore, and Arctic Engineering Division of ASME for publication in the JOURNAL OF OFFSHORE MECHANICS AND ARCTIC ENGINEERING. Manuscript received February 5, 2016; final manuscript received August 7, 2017; published online October 27, 2017. Assoc. Editor: Yi-Hsiang Yu.

J. Offshore Mech. Arct. Eng 140(2), 021901 (Oct 27, 2017) (8 pages) Paper No: OMAE-16-1012; doi: 10.1115/1.4037826 History: Received February 05, 2016; Revised August 07, 2017

## Abstract

Environmental conditions created by winds blowing oblique to the direction of the waves are necessary to conduct some survivability tests of offshore wind turbines. However, some facilities lack the capability to generate quality waves at a wide range of angles. Thus, having a wind generation system that can be rotated makes generating winds that blow oblique to the waves possible during survivability tests. Rotating the wind generation system may disrupt the flow generated by the fans because of the effect of adjacent walls. Closed or semiclosed wind tunnels may eliminate the issue of wall effects, but these types of wind tunnels could be difficult to position within a wave basin. In this work, a prototype wind generation system that can be adapted for offshore wind turbine testing is investigated. The wind generation system presented in this work has a return that minimizes the effect that the walls could potentially have on the fans. This study characterizes the configuration of a wind generation system using measurements of the velocity field, detailing mean velocities, flow directionality, and turbulence intensities. Measurements were taken downstream to evaluate the expected area of turbine operation and the shear zone. The dataset has aided in the identification of conditions that could potentially prevent the production of the desired flows. Therefore, this work provides a useful dataset that could be used in the design of wind generation systems and in the evaluation of the benefits of recirculating wind generation systems for offshore wind turbine research.

<>

## Introduction

The next step in development of renewable energy is in the field of offshore wind energy [1]. There is big potential for offshore wind energy that has been assessed in the United States [2]. For example, it has been calculated that there is a consistent wind density greater than 400 W/m2 at 20–50+ miles offshore of the U.S. northeast coast. At these potential wind energy sites, the depth of water usually makes the use of fixed-bottom monopile or jacket foundations economically unfeasible. Therefore, floating offshore wind turbines (FOWTs) and new technologies will need to be developed to make use of these sources of energy and to make offshore wind farms cost competitive with their terrestrial counterparts.

Modeling has been used to investigate the feasibility of utilizing new technologies in offshore wind power [35]. Design optimization is often performed through numerical methods and simulations. However, these numerical methods need to be validated with experimental data. The validation of numerical methods is particularly important in the case of floater designs, which are subject to complex motions [6]. Floater designs can be classified into three categories: tension-leg platform (TLP), spar-buoy, and semisubmersible [7,8]. Tension-leg platform based floating wind turbines have been studied and are in various stages of testing and development with significant advances already made by companies such as PelaStar, Seattle, WA, IBERDROLA (Biscay, Spain) (TLPWIND), and GICON (Dresden, Germany). Similarly, spar-buoy designs, namely, the Hywind design by Statoil (Stavanger, Norway), have been developed. One example of a semisubmersible design is the WindSea design, which is still in the early stages of development. Like many semisubmersibles, WindSea uses a trifloater design. Principle power has also designed a semisubmersible, the WindFloat. A full-scale prototype of the WindFloat has been successfully deployed off the coast of Portugal. The University of Maine installed a semisubmersible design, VolturnUS, off the coast of Maine in 2014. VolturnUS is a 1/8 scale prototype of a 6 MW commercial design. Each of these floater designs has its benefits and shortcomings when used to support different types of wind turbines.

Facilities located in several countries around the world possess a wind tunnel or wind generation system and a wave basin as testing grounds for scaled FOWT models. Wind generation systems used to generate airflow have been constructed at facilities such as the Maritime Research Institute Netherlands (MARIN) [9], OCEANIDE, Ecole Centrale de Nantes [10], IH Cantabria, IFREMER [11], Technical University of Denmark [12], and the Norwegian Marine Technology Research Institute [13] to test FOWT models. Notable efforts to use a wind generation system to test scaled offshore floating wind turbines are being made by MARIN and Ecole Central de Nantes. MARIN has developed a wind generation system with improved wind flow quality [9]. This wind generation system, initially developed for the DeepCwind Consortium, has been used in several testing campaigns. MARIN’s wind generation system generates wind environments with low swirl and an average turbulence of less than 5% intensity in the flow field. These wind characteristics were accomplished with a bank of 35 fans with a honeycomb front plate to reduce swirl and a nozzle to reduce turbulence [7]. This wind generation system was used in the 2011 and 2013 testing campaigns by the University of Maine at MARIN. Tests were conducted to evaluate different floating turbine designs such as the spar, TLP, and semisubmersible variants [14]. The three different floating systems were used in conjunction with the 1/50 scale National Renewable Energy Laboratory (NREL) 5 MW reference turbine in a wind-wave basin. These tests required that the turbine be exposed to a swirl-free flow with a turbulence intensity of about 5%. Multidirectional sea conditions and the ability to produce simultaneous stochastic wind and waves were also a requirement of testing [14]. This wind generation system, however, had some shortcomings. The bank of fans needed to be placed high enough to prevent it from interacting with the water. Interaction with the water would result in decreased wind speed on the lower portion of the rotor. The reduction of the flow near the waterline, caused by the increase in the height of the fans, was mitigated by tilting the fans approximately 2 deg downward. Repositioning the fans improved the wind speeds at the bottom of the rotor, but inadvertently introduced a vertical component to the wind velocity. The development of this wind generation system yielded valuable insights into the interaction of the wind generator with the building in which it was housed. The most observable effect of the facility walls on the performance of the wind system is that the fans required special attention due to the recirculation of the wind field in the basin and the variation of the wind speed with the distance from the fans [15].

Ecole Centrale de Nantes has developed a wind generation system that utilizes centrifugal fans instead of axial fans to avoid the generation of a twisted flow, which introduces spatial inhomogeneity and high turbulence levels. Additional steps were taken to reduce turbulence, increase homogeneity, and improve the quality of flow based on proven wind-generation system practices that included the use of a screen and a honeycomb. The evaluation of the wind generation system showed homogeneity of the average velocity in the test area that met the design requirements and a turbulence level of 3% at the center of the jet. All these results prove the relevance of different components used to reduce turbulence and to homogenize the flow and highlight the measures that a facility can take to produce, tune, test, and assure high-quality wind [10].

To design better FOWT systems, numerical results need to be validated. It has been found that some of the major loadings on the floating structures and the moorings occur at complex sea states when the wind does not align with the direction of the waves [16]. To generate complex sea states such as wind loading in directions that differ from that of the waves, the wind generation system has to be rotated. Successful rotation of a wind generation system above a wave basin is contingent on the ability of the tunnel to be insensitive to the different boundary conditions imposed by the building.

A prototype of the wind generation system (shown in Fig. 1) was built to evaluate the generated flow quality. This prototype was built to a one-third scale of the one installed at a new wind wave facility at the University of Maine. The scaled wind generation system has a 180-deg turn meant to reduce the effect of the walls on the fans. The objective of studying the wind generation system is to reproduce a variety of scaled real-world wind conditions including uniform flow and variation in the flow profile. The scaled wind generation system has been designed to be able to blow wind for a 1/130th scaled model of NREL’s 5 MW turbine. When testing wind turbines, the flow needs to be highly uniform throughout the test area and have low turbulent intensity. The target set for performance specifications for the prototype scaled wind generation system is a flow homogeneity of 5% or less, turbulence intensity less than 4%, and wind speeds up to that of a scaled hurricane. The prototype wind generation system was designed to generate a mean velocity output of 5 m/s and was housed in a building 30.48 m long, 18.29 m wide, and 4.26 m tall. This building also housed a shared laboratory, in which a 14 -m long and 5-m wide area at the middle of the building was designated exclusively for the operation of the wind generation system. When scaling the findings of this study to the full-scale wind generation system, Froude scaling will be utilized for the global parameters of the wind generation system [17] where elements such as honeycomb and screen will follow scaling methods described in Refs. [18,19].

## Theory

To evaluate the flow quality, the uniformity and directionality of the flow and the turbulence intensity were calculated. The measured flow is decomposed into a mean and turbulent component Display Formula

(1)$u(t)=u¯+u′(t)v(t)=v¯+v′(t)w(t)=w¯+w′(t)Ul(t)=(u(t)2+v(t)2+w(t)2)1/2$

where $u(t)$, $v(t)$, and $w(t)$ are the component flow measurements, $Ul$ (t) is the local combined flow measurement, $u$, $v$, and $w$, are the mean components of the flow and $u′(t)$, $v′(t)$, and $w′(t)$ are the turbulent components in the x, y, and z coordinates, respectively. The mean velocity, $Ul¯$, is calculated as Display Formula

(2)$U¯¯l=1N∑i=1NUi(N)$

where $N$ is the number of samples

The turbulent strength, $urms$, and turbulent intensity, T.I., are then calculated as Display Formula

(3)$urms=1N∑i=1N(ui′)2$
Display Formula
(4)$T.I.=urms/Ul¯$

where $Ul¯$ is the mean flow at the same location, and $N$ is the number of samples.

The directionality, $Dir$, is defined as Display Formula

(5)$Dir=u¯Ul¯$

and the error in flow direction is defined as [11] Display Formula

(6)$EDir=v¯2+w¯2Ul¯$

## Test Setup

Ten axial fans (280 W each) powered the prototype wind generation system by forcing air into individual square ducts that carried the air to a 180-deg turn before combining the flows in the settling chamber (Fig. 2). Axial fans were chosen over centrifugal fans to produce high volume wind in a generation system where significant pressure drops were not expected. The settling chamber was designed with slots that allowed frames of the same cross section to be slid in from the side of the chamber through an access panel. The air first passed through a section of honeycomb and then encountered the screen(s) further downstream. The nozzle reduced the rectangular cross section exiting the settling chamber to a final section 1.8 m wide × 1.2 m high with fiberglass flow restrictions further reducing the upper corners to a fileted radii of 0.5 m as shown in Fig. 2. This configuration results in a 41% reduction in the cross-sectional area equal to a contraction ratio of 1.7, an approximately 69% increase in mean flow rate and a 28.4% reduction in turbulence [20].

The honeycomb and screens that were installed in the settling chamber were employed to condition the flow while minimizing pressure losses across each device. The initial configuration of the wind generation system included the use of a 7.6 cm thick honeycomb with 1.25 cm wide cells shortly after the convergence of the ten square ducts. Immediately downstream from the honeycomb, a heavy gauge screen was installed as the structural support for a fine mesh screen (wire diameter of 1.52 mm and 3.05 mm opening) that was subsequently added one layer at a time to analyze the effectiveness of each layer. This structural steel mesh supported the screens along their entire span to avoid extreme deformation of the screens due to the drag forces they experienced.

The flow field produced by the wind generation system was measured by surveying 60% of the air jet, using an acoustic and hot wire anemometer (shown in Fig. 3) that were arranged to sample the flow. Maps of the mean air speed and turbulence intensity were produced with the data acquired. The maps of mean flow and turbulence intensity were created by moving the anemometers perpendicularly in and out of the test jet at ten different elevations, each 160 mm apart from one another in the vertical plane. The approximate locations of the ten different elevations are shown in Fig. 4 (labeled as measurements 1–10 in the figure). The hot wire was placed 10 mm behind the midline of the testing volume of the acoustic anemometer and was mounted with the element in a vertical position. In this position, the hotwire was most sensitive to turbulence intensity within the horizontal plane where the nozzle contraction was the greatest. A traverse (shown in Fig. 3) was used to move the acoustic anemometer and hot wire horizontally into the jet while measuring the velocity field. The traverse speed of 38.8 mm/s or 0.78% of the maximum recorded wind velocity in the test volume made any contributions to cross-flow components of air velocity negligible. The elevation of the linear traverse was then changed, and data was collected as the sensors were withdrawn from the testing volume. Ten different elevations within this plane were mapped to complete one cross-sectional measurement. These planar surveys (unless otherwise noted) were taken at 0.5, 1.0, 1.5, and 2.0 m from the nozzle of the wind generation system. Although the fans of the generation system can be operated at different rotational speeds, all the results presented in this work were obtained by operating the fans at the same rotational speed.

The acoustic anemometer that was used to measure the mean velocity throughout the test area was an R. M. Young Model 81000. This instrument can measure three-dimensional velocities of up to 40 m/s with an accuracy of ±0.05 m/s. The flow directionality was calculated using the flow angles measured by the acoustic anemometer. This device was configured to collect data at a rate of 32 Hz. A Dantec dynamics 55P01 wire probe hot wire anemometer was set to measure flow at a sampling rate of 5 kHz. To determine the turbulence intensity in the test area, the data from the hot wire anemometer was filtered to 2 kHz to match the sample rate of the data acquisition system. The turbulence intensity in a particular area is calculated using sets of 400 acquired data points in Eqs. (4) and (5).

## Results and Discussion

###### Flow Uniformity and Turbulence Intensity.

The turbulence created by the system has to be mitigated by the use of honeycomb, several layers of mesh screen, and a nozzle. Although it is known that adding more screens and honeycomb diminishes the turbulence, doing so adversely affects the maximum flow speed generated by the wind generation system [21]. The first sets of tests were, therefore, done to evaluate the influence of incrementing the number of screens on the turbulence and flow.

Figures 57 show the results at 0.5 m in front of the nozzle for flow and turbulence intensity using one, two, and three screens, respectively. The projected area of the nozzle is shown in the black dashed line in the figures. It can be seen in the surveys above that as each layer of screen was added, the variability in the velocity of the central testing area decreased. The airspeed throughout the testing area became more consistent at the expense of the maximum achievable velocity when the number of screens was increased. In the case in which one screen was used, the velocity at the center of the generation system was 6.5 m/s, and the turbulence intensity at the center of the nozzle was 0.0387. The addition of a second screen decreased the velocity in the center of the test area to 5.5 m/s and the turbulence intensity to 0.0281. A third screen reduced the mean velocity in the center of the test area further to 5.0 m/s, and the turbulence intensity to 0.0149 at the center of the nozzle. The application of three screens achieved the initial design requirements of maintaining low turbulence intensity while still being capable of producing the required maximum mean velocity output of 5 m/s.

Figures 710 were analyzed to determine the evolution of the flow field from the nozzle. These figures show the results of the survey throughout the test area starting at 0.5 m downstream from the nozzle, shown in Fig. 7, and moving an additional 0.5 m in each step until a distance of 2.0 m is reached (Fig. 10). The survey results acquired at 0.5, 1.0, 1.5, and 2.0 m were taken with the fans operating at the same rotational speed. In the analysis of the data, the evolution of the shear zone and the contraction of the region with low turbulence were most apparent. As the air from the generation system moved farther downstream from the nozzle, the thickness of the shear zone increased. With the evolution and expansion of the shear zone, the vorticity dissipated causing the measured turbulence intensity to decrease.

In Fig. 7, it can be seen that at 0.5 m from the nozzle, an area of low-velocity flow was produced in the upper portion of the midline of the generation system. In Figs. 710, the deficiency in flow at 0.5 m is shown to recover by the time the jet travels 2.0 m from the nozzle. The flow deficiency in proximity to the nozzle could be due to the separation of the airflow after it passes the fan and the small inner radius of the U-turn section. Any contribution of the small radius of the U-turn to the airflow profile is expected to be mitigated in the full-scale wind generation system [22]. Future tuning of this wind generation system will include the use of a diffuser between each fan and the U-turn.

A second dataset was acquired with the wind generation system generating 4 m/s at the midpoint of the nozzle. Figures 1114 represent the survey of the wind generation system with honeycomb and three screens. The mean flow and turbulence intensity are shown in Figs. 1114 starting at 0.5 m from the nozzle (Fig. 11) and moving downstream in 0.5 -m increments to 2.0 m from the nozzle (Fig. 14). There was a large field of homogeneous flow that extended to within 0.10–0.20 m of the nozzle projection, and turbulence intensity was at an acceptable level of less than 1.5% at 0.5 m from the nozzle. The shear zone expanded, and the turbulence strength decreased as measurements were taken further downstream as was seen in the dataset acquired at a flow of 5 m/s at the nozzle midpoint. It can also be seen that there was no significant effect of the fan suction on the quality of the flow from 0.5 to 2.0 m.

###### Flow Directionality.

Figures 1517 show the effect of the number of screens used on flow directionality in the wind generation system. Figure 15 shows the error in the flow direction of the air flow generated by the wind generation system when one screen was used 0.5 m from the nozzle. It can be seen that there was a large area located at 0.9–1.2 m in the vertical and 0.0–0.3 m in the horizontal where the error in flow direction was large, up to 43%. Figure 16 shows the error in the flow direction when the number of screens was increased to two at 0.5 m from the nozzle. It can be seen that the error in flow direction dramatically decreased up to 3.5% at a zone located around 1 m vertically and 0 m horizontally. However, there was a zone around 0.5 m in the horizontal and 0.2 in the vertical, where there was an error in the flow direction up to 18%. Figure 17 shows the error in the flow direction when three screens were used in the wind generation system at 0.5 m from the nozzle. It can be seen that the error in the flow direction in the zone around 0.5 m in the horizontal and 0.2 in the vertical was slightly diminished. The error in flow direction generated by the wind generation system equipped with three screens is shown in Fig. 18. The flow was measured at 1.0 m from the nozzle. It can be seen that the error in flow direction is smoothened in the test area to less than 9%. Similarly, Figs. 19 and 20 show the error in flow directions generated by the wind generation system at 4 m/s. The flow was measured at 1.5 and 2.0 m from the nozzle. The error in the flow direction in the zone around 0.5 m in the horizontal and 0.2 in the vertical direction was slightly higher in the flow at 4.0 m/s than it was at 5.0 m/s. Figure 19 shows that there was an error in the flow direction up to 18% around a zone located at 0.5 m in the horizontal and 0.2 in the vertical. Figure 20 shows that the error in the flow direction is only reduced to 17% in some areas. Although there was no specific requirement for the maximum error in flow directionality, this error was sought to be kept below 9% (5 deg). For this reason, this area of high error in flow directionality (17–18%) will be studied with individual control of the fans and turning vanes and screens in the future.

###### Dynamic Response.

The wind generation system response was evaluated by measuring the wind speed generated by the wind generation system due to a commanded wind condition. The evaluation of the dynamic response is important since it provides insight into the maximum variation in the flow at a given frequency. The flow measurements were acquired 1 m from the nozzle at the projected center of the nozzle. This location was selected since it is representative of the flow that would go through a turbine. Two different wind speed profiles were used to show the wind generation system response. The first wind speed profile is shown in Fig. 21. In this wind speed profile, the frequency of the prescribed wind condition was continuously increased, the mean air speed was 3.2 m/s, and the amplitude was 3.2 m/s. The wind generation system produced wind speed oscillations with lower amplitudes as the airflow frequency increased, particularly at frequencies above 0.083 Hz (the maximum acceleration at this frequency and amplitude is around 1.78 m/s2). The maximum frequency at which the wind generation system produced oscillating airflow was around 0.9 Hz. In the second profile presented in Fig. 22, the frequency of the prescribed wind speed was progressively increased, the mean air speed was 4.5 m/s, and the amplitude was 1.7 m/s. At frequencies above 0.167 Hz (the maximum acceleration at this frequency and amplitude is around 1.67 m/s2), the generated wind speed decreased as the frequency of the wind oscillations was increased. The maximum frequency at which the wind generation system generates airflow also appeared to be around 0.9 Hz as it was in the previous tests. From the results of these two tests, it could be inferred that the maximum flow acceleration was around 1.6–1.8 m/s2.

## Conclusions

This paper describes testing performed to evaluate the flow quality generated by an open jet wind generation system with a 180 deg turn. The first objective of the testing was to evaluate the uniformity of the flow, turbulence intensity, and flow direction. The experimental data show that the three screens and one honeycomb configuration provide adequate turbulence attenuation and flow direction. This configuration achieves the design requirement of moderate turbulence while producing airflow at 5 m/s. The experimental dataset also shows that there is no significant increase in turbulence in the turbine test area due to the evolution of the shear zone up to 2 m. Thus, the dataset has provided information on the area in which the turbine can be operated with a uniform and low turbulent flow. The dataset also provided information on areas where the error in flow directionality needs to be reduced. This information is valuable when testing a scaled wind turbine that is subject to roll motions. The experimental dataset also shows that there is no significant impact of the suction flow on the quality of the flow downstream and the wind generation system could benefit from the addition of diffusers after each fan.

The presented dataset will be helpful when evaluating other wind generation systems, such as the semiclosed wind tunnel. The presented dataset will be used to compare the quality of the flow when a wall is present at a certain distance from the nozzle. Additionally, this dataset could be critical to the design of the collector and the U-return section of the semiclosed wind tunnel. Baseline flow and turbulence data collected from the current configuration of the wind tunnel will be used to identify the effect of the U-return and diffuser installation in the future.

The experimental data presented in this paper offer a dataset that can be used to aid in the design of wind generation systems for offshore wind turbine research. The experimental data also provide insight into flow directionality and turbulence intensity as a function of distance from the nozzle, which is important to know in order to correctly place the scaled wind turbine. The goal of this work was to provide a dataset that could be used as a benchmark to evaluate the benefits of recirculating wind generation systems for offshore wind turbine research. Future work will focus on varying the fan speeds to improve the quality of the flow produced by the wind generation system.

## Funding Data

• The U.S. National Science Foundation (Grant No. 1337895).

• EDA and the University of Maine.

## Nomenclature

• $Dir$ =

directionality

• $EDir$ =

error in flow direction

• $N$ =

number of samples

• $u¯$ =

mean components of the flow in the x coordinate

• $u(t)$ =

flow measurement in the x coordinate

• $urms$, =

turbulent strength

• $Ul$ (t) =

local combined flow measurement

• $u′(t)$ =

turbulent component in the x coordinate

• $Ul¯$ =

local mean flow

• $v¯$ =

mean components of the flow in the y coordinate

• $v(t)$ =

flow measurement in the y coordinate

• $v′(t)$ =

turbulent component in the y coordinate

• $w¯$ =

mean components of the flow in the z coordinate

• $w(t)$ =

flow measurement in the z coordinate

• $w′(t)$ =

turbulent component in the z coordinate

## References

Heronemus, W. E. , 1972, “ Pollution-Free Energy From Offshore Winds,” Eighth Annual Conference and Exposition Marine Technology Society, Washington, DC, Sept. 11–13.
Musial, W. D. , and Butterfield, C. P. , 2004, “ Future for Offshore Wind Energy in the United States,” Energy Ocean Conference, Palm Beach, FL, June 28–29, pp. 4–6.
Chakrabarti, S. K. , 1994, Offshore Structure Modeling, World Scientific Publishing, Singapore, Chap. 7.
Jonkman, J. M. , 2007, “ Dynamics Modeling and Loads Analysis of an Offshore Floating Wind Turbine,” National Renewable Energy Laboratory, Golden, CO, Report No. NREL/TP-500-41958.
Jonkman, J. M. , and Sclavounos, P. D. , 2006, “ Development of Fully Coupled Aeroelastic and Hydrodynamic Models for Offshore Wind Turbines,” National Renewable Energy Laboratory, Golden, CO, Technical Report No. NREL/CP-500-39066.
Browning, J. R. , Jonkman, J. , Robertson, A. , and Goupee, A. J. , 2014, “ Calibration and Validation of a Spar-Type Floating Offshore Wind Turbine Model Using the FAST Dynamic Simulation Tool,” J. Phys.: Conf. Ser., 555(1), p. 012015.
Roddier, D. , Cermelli, C. , Aubault, A. , and Weinstein, A. , 2010, “ WindFloat: A Floating Foundation for Offshore Wind Turbines,” J. Renewable Sustainable Energy, 2(3), p. 033104.
Butterfield, S. , Musial, W. , Jonkman, J. , Sclavounos, P. , and Wayman, L. , 2007, “ Engineering Challenges for Floating Offshore Wind Turbines,” National Renewable Energy Laboratory, Golden, CO, Report No. NREL/CP-500-38776.
de Ridder, E. , Otto, W. , Zondervan, G. , Savenije, F. , and Huijs, F. , 2013, “ State of the Art Model Testing Techniques for Floating Wind Turbines,” EWEA Offshore, Frankfurt, Germany, Nov. 19–21.
Courbois, A. , Flamand, O. , Toularastel, J.-L. , Ferrant, P. , and Rousset, J.-M. , 2011, “ Applying Relevant Wind Generation Techniques to the Case of Floating Wind Turbines,” Sixth European and African Conference on Wind Engineering (EACWE), Nantes, France, July 7–13, pp. 1–8.
Ohana, J. , Le Boulluec, M. , Peron, E. , Klinghammer, C. , Tancray, A. , and Mansuy, E. , 2014, “ Open Jet Blower Type Wind Generator With Variable Wind Speed Capability for Physical Model Testing of Offshore Structures,” Fifth International Conference on the Application of Physical Modelling to Port and Coastal Protection (Coastlab14), Varna, Bulgaria, Sept. 29–Oct. 2.
Bredmose, H. , Mikkelsen, R. , Hansen, A. M. , Laugesen, R. , Heilskov, N. , Jensen, B. , and Kirkegaard, J. , 2015, “ Experimental Study of the DTU 10 MW Wind Turbine on a TLP Floater in Waves and Wind,” EWEA Offshore Conference, Copenhagen, Denmark, Mar. 10.
Nielsen, F. G. , Hanson, T. D. , and Skaare, B. , 2006, “ Integrated Dynamic Analysis of Floating Offshore Wind Turbines,” ASME Paper No. OMAE2006-92291.
Kimball, R. W. , Goupee, A. J. , Coulling, A. J. , and Dagher, H. J. , 2012, “ Model Test Comparisons of TLP, Spar-Buoy and Semi-Submersible Floating Offshore Wind Turbine Systems,” SNAME Annual Meeting and Expo, Providence, RI, Oct. 24–26, pp. 24–26.
Robertson, A. N. , Jonkman, J. M. , Masciola, M. D. , Molta, P. , Goupee, A. J. , and Coulling, A. J. , 2013, “ Summary of Conclusions and Recommendations Drawn From the DeepCWind Scaled Floating Offshore Wind System Test Campaign,” National Renewable Energy Laboratory, Golden, CO, Report No. NREL/CP-5000-58076.
Barj, L. , Stewart, S. , Stewart, G. , Lackner, M. , Jonkman, J. , and Robertson, A. , 2014, “ Wind/Wave Misalignment in the Loads Analysis of a Floating Offshore Wind Turbine,” AIAA Paper No. 2014-0363.
Navqvi, S. K. , 2012, “ Scale Model Experiments on Floating Offshore Wind Turbines,” Ph.D. dissertation, Worcester Polytechnic Institute, Worcester, MA.
Bradshaw, P. , and Mehta, R. D. , 1979, “ Design Rules for Small Low Speed Wind Tunnels,” Aeronaut. J. R. Aeronaut. Soc., 83(827), pp. 443–449.
Farell, C. , and Youssef, S. , 1992, “ Experiments on Turbulence Management Using Screens and Honeycombs,” University of Minnesota, Minneapolis, MN, Project Report No. 338.
Batchelor, G. K. , 1953, The Theory of Homogeneous Turbulence, Cambridge University Press, Cambridge, UK, pp. 55–75.
Barlow, J. , Rae, W. , and Pope, A. , 1999, Low Speed Wind Tunnel Testing, Wiley, New York, Chap. 2.
McDonalogue, D. J. , and Srivastava, R. S. , 1968, “ Motion of a Fluid in a Curved Tube,” Proc. R. Soc. A, 307(1488), pp. 37–57.
View article in PDF format.

## References

Heronemus, W. E. , 1972, “ Pollution-Free Energy From Offshore Winds,” Eighth Annual Conference and Exposition Marine Technology Society, Washington, DC, Sept. 11–13.
Musial, W. D. , and Butterfield, C. P. , 2004, “ Future for Offshore Wind Energy in the United States,” Energy Ocean Conference, Palm Beach, FL, June 28–29, pp. 4–6.
Chakrabarti, S. K. , 1994, Offshore Structure Modeling, World Scientific Publishing, Singapore, Chap. 7.
Jonkman, J. M. , 2007, “ Dynamics Modeling and Loads Analysis of an Offshore Floating Wind Turbine,” National Renewable Energy Laboratory, Golden, CO, Report No. NREL/TP-500-41958.
Jonkman, J. M. , and Sclavounos, P. D. , 2006, “ Development of Fully Coupled Aeroelastic and Hydrodynamic Models for Offshore Wind Turbines,” National Renewable Energy Laboratory, Golden, CO, Technical Report No. NREL/CP-500-39066.
Browning, J. R. , Jonkman, J. , Robertson, A. , and Goupee, A. J. , 2014, “ Calibration and Validation of a Spar-Type Floating Offshore Wind Turbine Model Using the FAST Dynamic Simulation Tool,” J. Phys.: Conf. Ser., 555(1), p. 012015.
Roddier, D. , Cermelli, C. , Aubault, A. , and Weinstein, A. , 2010, “ WindFloat: A Floating Foundation for Offshore Wind Turbines,” J. Renewable Sustainable Energy, 2(3), p. 033104.
Butterfield, S. , Musial, W. , Jonkman, J. , Sclavounos, P. , and Wayman, L. , 2007, “ Engineering Challenges for Floating Offshore Wind Turbines,” National Renewable Energy Laboratory, Golden, CO, Report No. NREL/CP-500-38776.
de Ridder, E. , Otto, W. , Zondervan, G. , Savenije, F. , and Huijs, F. , 2013, “ State of the Art Model Testing Techniques for Floating Wind Turbines,” EWEA Offshore, Frankfurt, Germany, Nov. 19–21.
Courbois, A. , Flamand, O. , Toularastel, J.-L. , Ferrant, P. , and Rousset, J.-M. , 2011, “ Applying Relevant Wind Generation Techniques to the Case of Floating Wind Turbines,” Sixth European and African Conference on Wind Engineering (EACWE), Nantes, France, July 7–13, pp. 1–8.
Ohana, J. , Le Boulluec, M. , Peron, E. , Klinghammer, C. , Tancray, A. , and Mansuy, E. , 2014, “ Open Jet Blower Type Wind Generator With Variable Wind Speed Capability for Physical Model Testing of Offshore Structures,” Fifth International Conference on the Application of Physical Modelling to Port and Coastal Protection (Coastlab14), Varna, Bulgaria, Sept. 29–Oct. 2.
Bredmose, H. , Mikkelsen, R. , Hansen, A. M. , Laugesen, R. , Heilskov, N. , Jensen, B. , and Kirkegaard, J. , 2015, “ Experimental Study of the DTU 10 MW Wind Turbine on a TLP Floater in Waves and Wind,” EWEA Offshore Conference, Copenhagen, Denmark, Mar. 10.
Nielsen, F. G. , Hanson, T. D. , and Skaare, B. , 2006, “ Integrated Dynamic Analysis of Floating Offshore Wind Turbines,” ASME Paper No. OMAE2006-92291.
Kimball, R. W. , Goupee, A. J. , Coulling, A. J. , and Dagher, H. J. , 2012, “ Model Test Comparisons of TLP, Spar-Buoy and Semi-Submersible Floating Offshore Wind Turbine Systems,” SNAME Annual Meeting and Expo, Providence, RI, Oct. 24–26, pp. 24–26.
Robertson, A. N. , Jonkman, J. M. , Masciola, M. D. , Molta, P. , Goupee, A. J. , and Coulling, A. J. , 2013, “ Summary of Conclusions and Recommendations Drawn From the DeepCWind Scaled Floating Offshore Wind System Test Campaign,” National Renewable Energy Laboratory, Golden, CO, Report No. NREL/CP-5000-58076.
Barj, L. , Stewart, S. , Stewart, G. , Lackner, M. , Jonkman, J. , and Robertson, A. , 2014, “ Wind/Wave Misalignment in the Loads Analysis of a Floating Offshore Wind Turbine,” AIAA Paper No. 2014-0363.
Navqvi, S. K. , 2012, “ Scale Model Experiments on Floating Offshore Wind Turbines,” Ph.D. dissertation, Worcester Polytechnic Institute, Worcester, MA.
Bradshaw, P. , and Mehta, R. D. , 1979, “ Design Rules for Small Low Speed Wind Tunnels,” Aeronaut. J. R. Aeronaut. Soc., 83(827), pp. 443–449.
Farell, C. , and Youssef, S. , 1992, “ Experiments on Turbulence Management Using Screens and Honeycombs,” University of Minnesota, Minneapolis, MN, Project Report No. 338.
Batchelor, G. K. , 1953, The Theory of Homogeneous Turbulence, Cambridge University Press, Cambridge, UK, pp. 55–75.
Barlow, J. , Rae, W. , and Pope, A. , 1999, Low Speed Wind Tunnel Testing, Wiley, New York, Chap. 2.
McDonalogue, D. J. , and Srivastava, R. S. , 1968, “ Motion of a Fluid in a Curved Tube,” Proc. R. Soc. A, 307(1488), pp. 37–57.

## Figures

Fig. 1

The different sections of the prototype wind-tunnel with recirculation are shown

Fig. 2

Front view of the wind generation system showing the ten fan arrangement and nozzle

Fig. 3

The acoustic and hot wire anemometers arrangement mounted on the traverse is shown

Fig. 4

The ten elevations relative to the nozzle at which measurements were taken are shown

Fig. 5

Flow field at 0.5 m from the nozzle using one screen

Fig. 6

Flow field at 0.5 m from the nozzle using two screens

Fig. 7

Flow field at 0.5 m from the nozzle using three screens with a 5 m/s flow at the center point

Fig. 8

Flow field at 1.0 m from the nozzle using three screens with a 5 m/s flow at the center point

Fig. 9

Flow field at 1.5 m from the nozzle using three screens with a 5 m/s flow at the center point

Fig. 10

Flow field at 2.0 m from the nozzle using three screens with a 5 m/s flow at the center point

Fig. 11

Flow field at 0.5 m from the nozzle using three screens with a 4 m/s flow at the center point (no turbine)

Fig. 12

Flow field at 1.0 m from the nozzle using three screens with a 4 m/s flow at the center point (no turbine)

Fig. 13

Flow field at 1.5 m from the nozzle using three screens with a 4 m/s flow at the center point (no turbine)

Fig. 14

Flow field at 2.0 m from the nozzle using three screens with a 4 m/s flow at the center point (no turbine)

Fig. 15

Flow directionality and error in flow direction at 0.5 m from the nozzle using one screen

Fig. 16

Flow directionality and error in flow direction at 0.5 m from the nozzle using two screens

Fig. 17

Flow directionality and error in flow direction at 0.5 m from the nozzle using three screens at 5 m/s

Fig. 18

Flow directionality and error in flow direction at 2.0 m from the nozzle using three screens at 5 m/s

Fig. 19

Flow directionality and error in flow direction at 0.5 m from the nozzle using three screens at 4 m/s

Fig. 20

Flow directionality and error in flow direction at 2.0 m from the nozzle using three screens at 4 m/s

Fig. 21

Wind generation system response to a prescribed wind speed profile with different frequencies (mean flow 3.2 m/s)

Fig. 22

Wind generation system response to a prescribed wind speed profile with different frequencies (mean flow 4.5 m/s)

## Discussions

Some tools below are only available to our subscribers or users with an online account.

### Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related Proceedings Articles
Related eBook Content
Topic Collections