Renewable Energy 118 (2018) 1e13

Contents lists available at ScienceDirect

Renewable Energy
journal homepage: www.elsevier.com/locate/renene

Experimental investigation of vertical-axis wind-turbine wakes in

boundary layer flow
-Agel*
Vincent F-C. Rolin, Fernando Porte
Ecole Polytechnique F
ed
erale de Lausanne (EPFL), EPFL-ENAC-IIE-WIRE, Lausanne, 1015, Switzerland

a r t i c l e i n f o a b s t r a c t

Article history: In this experiment, a small scale vertical-axis wind-turbine (VAWT) is immersed in a boundary-layer in a
Received 3 April 2017 wind tunnel and stereo particle image velocimetry is employed to quantify the 3D characteristics of the
Received in revised form wake. The measurements show that the wake is strongest behind the sector of the rotor which turns into
29 September 2017
the wind. Two counter-rotating vortex pairs in the wake induce crosswind motion which reintroduces
Accepted 29 October 2017
streamwise momentum into the wake. Terms of the mean kinetic energy budget are computed and
Available online 1 November 2017
demonstrate that this crosswind flow has a significant influence on the redistribution of momentum in
the wake. A similar analysis of the turbulence kinetic energy budget identifies shearing at the boundary
Keywords:
Atmospheric boundary layer
of the wake as the primary contributor to the production of turbulence. An analytical model is developed
Vertical-axis wind-turbine in order to obtain a theoretical basis from which to understand how the aerodynamic behavior of VAWTs
Wind-turbine wakes induces crosswind motion consistent with the production of counter-rotating vortex pairs.
Stereo-PIV © 2017 Elsevier Ltd. All rights reserved.
Wind tunnel

1. Introduction technology in order to determine the direction of the incoming

wind and techniques to forecast how it will change in the future. An
Carbon emissions produced by the consumption of fossil fuels additional benefit of VAWTs is that, due to their design, the
are a major contributor to climate change, and dependence on oil generator can be placed below the rotor which reduces loading on
combined with the volatility of oil prices is a growing source of the tower relative to HAWTs. This configuration presents a poten-
geopolitical tension. As a result, renewable energy has become an tial cost saving for the construction, installation, and maintenance
important field of study due to its demonstrated potential as a of the turbine.
sustainable, emission-free alternative to fossil fuels. Wind power is The geometry of VAWTs may also prove advantageous to plan-
one of the fastest growing and most cost-efficient of the different ning a windfarm. One way a rotor can capture wind energy is by
means of producing renewable energy available today. The largest presenting a larger frontal area to the wind. HAWTs can enlarge this
horizontal-axis wind-turbines (HAWTs) are able to generate 6 MW area by having a larger diameter, but this means that the lateral
of power and are able to extract close to the theoretically maximum distance between HAWTs must be increased to accommodate the
energy available from the wind, making them an appealing tech- larger rotor. VAWTs, by contrast, can expand the frontal area they
nology for large-scale energy production. Vertical-axis wind-tur- present to the wind by increasing the height of the rotor, rather
bines (VAWTs), however, offer certain advantages over HAWTs than the turbine diameter. This suggests that tall VAWTs can be
which merit study. They have, nonetheless, been the subject of placed closely together to conserve space and increase the power
significantly less research. density of a wind farm. Dabiri [1] demonstrated the potential
One advantage of VAWTs is that the power they generate is advantage of a cluster closely spaced group of counter-rotating
independent of the direction of the incoming wind as the turbine's VAWTs.
axis of rotation is oriented perpendicular to the flow. This is an These advantages have made VAWTs the center of an expanding
important advantage over HAWTs which have to face the wind in area of study, yet a vast gulf of knowledge still exists between our
order to attain peak performance. This requires sophisticated understanding of VAWT and HAWT wakes. Studies of HAWTs have
shown that a thorough characterization of the wake is critical to be
able to plan a turbine array for optimal performance and longevity
[2]. Specifically it is important to quantify the magnitude of tur-
* Corresponding author.
-Agel).
E-mail address: fernando.porte-agel@epfl.ch (F. Porte bulence, velocity deficit, and recovery rate of the wake behind

https://doi.org/10.1016/j.renene.2017.10.105
0960-1481/© 2017 Elsevier Ltd. All rights reserved.
2 V.F.-C. Rolin, F. Porte-Agel / Renewable Energy 118 (2018) 1e13

turbines upwind as it will influence those downwind. Under- opposed to a rotor driven by a motor, shows that the wake observed
standing how the wake is influenced by different inflow conditions in either case is indistinguishable so long as the tangential force
such as a boundary layer representative of atmospheric flow is produced by the interaction of the rotor and flow is consistent with
important for scaling experiments to reality and for validating the turbine's direction of rotation. Given this, the present study
numerical studies. Many previous experimental investigations have focuses on characterizing the wake of a small-scale motorized
been limited either by the implemented techniques which pro- VAWT in a boundary layer wind tunnel and how it develops
duced measurements of the wake with poor spatial resolution downstream. The mechanisms which govern the re-energization of
[3e5]. Other studies have been limited in scope, focused primarily the flow are identified and quantified via the terms of the mean
on characterizing the complex dynamics of blade-blade interaction, kinetic energy budget and momentum fluxes, and terms of the
vortex shedding, and the wake behavior only within the rotor and turbulence kinetic energy budget equation are computed in order
near wake regions [6e13]. The far wake, however, is the most to identify the mechanisms most responsible for the production of
relevant zone regarding turbine-turbine interactions in wind- turbulence in the wake. An analytical model is developed in order
farms. to develop a theoretical basis to understand the lateral forces
Of the few experimental studies which have investigated the far exerted on the flow believed to be the cause of the counter-rotating
wake, most have been limited to measurements over a single plane vortex pair observed in the wake.
[4,14] or did not fully immerse the turbine in the type of boundary
layer flow which would be encountered in a wind farm [15]. These
2. Experimental setup
studies have not reported consistent behavior of the wake but tend
to concur that the velocity deficit and turbulence in the wake are
Measurements were taken using stereo particle image veloc-
not symmetrically distributed behind the turbine and that the
imetry (stereo-PIV) behind a small scale H-type VAWT in the WIRE
structure and magnitude of the wake is largely dependent on the
boundary layer wind tunnel, depicted in Fig. 1.
rotor solidity and tip-speed ratio. Khosravi et al. [16] found that the
The wind tunnel is powered by a 130 kW fan which can generate
wake becomes more symmetric with increasing tip-speed ratio.
-Agel [14] observed peak winds up to 25 ms
1 . The test section is 28 m long with a cross-
Bergeles et al. [5] and Rolin and Porte
section that is 2:57 m wide and 2:00 m tall. Measurements are
levels of velocity deficit behind the sector of the turbine's rotation
conducted 20 m from the tunnel contraction such that a boundary
during which the blades turn into the wind. The highest levels of
layer develops naturally over the floor of the facility. Vertical pro-
turbulence in these studies were in the shear layer observed be-
files of the mean velocity and streamwise turbulence intensity of
tween this region and the unperturbed flow. Brochier et al. [4],
the incoming flow, acquired via hot-wire anemometry, are plotted
Battisti et al. [9] and Posa et al. [17], however, observed the highest
in Fig. 2.
turbulence levels in the wake behind the opposite side of the rotor
The depth of the boundary layer at the rotor position is 40 cm
where the blades turn with the wind. These observations attribute
the high turbulence in these regions to shear between the blades and the velocity in the free stream is 11 ms
1 . The relationship
and surrounding flow as well as the production of dynamic stall between velocity and height is roughly logarithmic up to height of
vortices. 18 cm. From the log-law approximation,
Some studies have observed phenomena which were not seen
U 1 z
elsewhere. Ryan et al. [15] reported the presence of a pair of ¼ ln ; (1)
u k z0
counter-rotating vortices in the wake of the VAWT but did not offer
an explanation for their presence. Bastankhah and Porte -Agel [18]
the values of the friction velocity and roughness are deduced as
observed a similar phenomenon in the wake of yawed HAWTs
u ¼ 0:44 ms
1 , and z0 ¼ 2:5  10
5 m respectively for k ¼ 0:4. This
and noted the similarity to the vortices associated with crossflow
approximation is presented in Fig. 3.
jets and concluded that the crosswind force exerted by the rotor
The rotor studied in this experiment is a three-bladed H-type
induces the conditions required to produce these vortices.
Darrieus as depicted in Fig. 4.
Previous experiments have investigated the validity of small-
The blades are NACA0018 airfoils with a chord length of 3 cm
scale empirical studies on VAWTs. Measurements performed over
and a span of 15:5 cm. The diameter of the rotor is 16:6 cm. The
a range of Reynolds numbers by Bachant and Wosnik [19] indicate
blockage ratio of the rotor frontal area to the tunnel cross-section is
that the wake characteristics are largely Reynolds insensitive. This
0:5%. The turbine operates at a tip-speed ratio of l ¼ 1:1 based on a
suggests that empirical observations at low Reynolds numbers are
scalable. A comparison of measurements by Araya and Dabiri [20] wind speed at the rotor mid-span of 9:43 ms
1 . Two Reynolds
performed behind a rotor driven by the incoming wind, as numbers are used to characterize the flow with respect to the
turbine, based on the incoming wind speed and rotor diameter,

Fig. 1. Schematic of the boundary layer wind tunnel of the WIRE laboratory.
 V.F.-C. Rolin, F. Porte-Agel / Renewable Energy 118 (2018) 1e13 3

Fig. 2. Profiles of the incoming boundary layer (left) velocity and (right) turbulence intensity. The rotor is illustrated to scale for reference.

ReD ¼ 100; 000, or on the airfoil chord-length and the speed rela- snapshots of the flow. The shape and structure of the wake is
tive to the blades, ReC ¼ 20; 000. The turbine is driven by a 12 W presented in Fig. 6, which shows mean velocity deficit over all
brushless Maxon EC45 flat motor in order to maintain a constant measured planes.
rotational velocity and level of performance. The motor was The velocity deficit at each individual measurement plane is
selected specifically for its low profile in order to keep the rotor presented in Fig. 7.
within the boundary layer and to avoid strong interactions with the Comparing the velocity deficit to the position of the rotor in-
wind. dicates that the far wake is deflected towards the negative-y di-
Stereo-PIV is implemented as illustrated in Fig. 5. rection, and recovers more momentum above the rotor midspan
A dual pulsed laser passes through a 10 mm cylindrical lens to than below. Fig. 8 presents a horizontal vertical profile of the ve-
illuminate a plane of interest across the wind-tunnel. Two cameras locity deficit, showing clearly that the velocity deficit in the wake
are used to obtain stereoscopic images of the plane, which is unevenly distributed such that the momentum behind the nega-
aligned normal to the streamwise direction. One camera is placed tive-y sector of the rotor which turns into the wind is significantly
upstream and the other downstream with respect to the measured lower than behind the positive-y sector. The vertical profile of ve-
flow, with an angle of roughly 87+ between them and a distance of locity deficit in Fig. 9 shows that the higher region of the wake
approximately 1:7 m from the center of the plane. This configura- returns to the inlet condition more quickly than the lower portion.
tion allows the major flow features to be captured over a 3D  4D The vorticity at each measurement plane is presented in Fig. 10
area. Post-processing is performed over 32  32 pixel interrogation with the in-plane velocity vectors overlaid.
windows with 50% overlap, resulting in 250  168 points of data This figure clearly indicates the presence of two pairs of counter-
with a resolution of 0:02D. The sampling frequency is roughly rotating vortices. The pairs are of different magnitudes, with those
10 Hz, with 200ms between exposures. Measurements are taken of in the wake of the negative-y sector of the rotor which turns into
the incoming flow and of the wake at 2, 3, 4, 6, 8, and 10 rotor di-
ameters from the turbine.

3. Results

In this section the data acquired by stereo-PIV are analyzed and

reported. Mean quantities are computed as time averages of

Fig. 3. The measured (dots) and logarithmic fit (solid line) of velocity in the boundary
layer. Fig. 4. Model H-type vertical-axis wind-turbine.
4 V.F.-C. Rolin, F. Porte-Agel / Renewable Energy 118 (2018) 1e13

Fig. 5. Top-down view of stereo-PIV configuration.

Fig. 6. Contours of streamwise velocity deficit normalized by the incoming wind speed at the midspan over each measurement plane, with the rotor situated at the origin. The
turbine's direction of rotation is indicated by a black arrow.

Fig. 7. Contours of streamwise velocity deficit at different downstream positions normalized by the incoming wind speed at the midspan. In-plane velocity vectors are overlaid. The
outline of the rotor is delineated by a black box.
 V.F.-C. Rolin, F. Porte-Agel / Renewable Energy 118 (2018) 1e13 5

Fig. 8. Crosswind profiles of the streamwise momentum deficit normalized by the incoming velocity at the midspan. Dotted lines denote the lateral edges of the VAWT rotor for
reference.

Fig. 9. Vertical profiles of streamwise velocity normalized by the incoming wind-speed at the midspan. The dashed line represents the incoming boundary layer flow. Dotted lines
denote the upper and lower edges of the VAWT rotor for reference.

Fig. 10. Contours of vorticity and in-plane velocity vectors at different downstream positions normalized by the incoming wind speed at the midspan and rotor diameter. In-plane
velocity vectors are overlaid. The outline of the rotor is delineated by a black box.
6 V.F.-C. Rolin, F. Porte-Agel / Renewable Energy 118 (2018) 1e13

Fig. 11. Contours of TKEþ at different downstream positions normalized by the incoming wind speed at the midspan. In-plane velocity vectors are overlaid. The outline of the rotor
is delineated by a black box.

the wind being significantly larger and stronger than those in the 3.1. Mean kinetic energy budget
wake of the positive-y sector. These vortices appear to induce sig-
nificant crosswind motion. The budget of mean kinetic energy, or MKE, is a useful tool for
The added turbulence kinetic energy, or TKE, is defined as the identifying phenomena responsible for the production and
difference between the TKE in the wake and the incoming destruction of kinetic energy. The budget equation decomposes the
boundary layer, TKEþ ¼ TKEw
TKE0 . Fig. 11 shows the magnitude conservation of MKE into components which each describe
of added TKE at each measurement plane. different mechanisms responsible for re-energizing the wake and
It is clear that the highest levels of TKE occur at the periphery of redistributing momentum. Expressed using Einstein index nota-
the wake where there is strong shear. Fig. 12 shows profiles of TKEþ tion, the equation for the mean kinetic energy budget is
which show that the stronger velocity gradient on the negative-y
boundary of the wake produces significantly stronger turbulence. 2 2
1 vU i 1 vU U vP
The vertical profile of TKE in the wake, shown in Fig. 13, shows that þ Uj i ¼
g di3 U i
i
the faster moving flow higher in the boundary layer causes a 2 vt 2 vx r vxj
|fflfflffl{zfflfflffl} |fflfflfflfflffl{zfflfflfflfflffl} |fflfflfflfflffl{zfflfflfflfflffl} |fflffl
j
ffl{zfflfflffl}
stronger velocity gradient at the upper boundary of the wake, I III
II IV
leading to stronger turbulence. (2)
The high velocity deficit close to the ground, and high turbu- v2 U vU vu0i u0j U i
þ nU i 2 i þ u0i u0j i
:
lence level near the top of the wake is consistent with the previ- vxj vxj vxj
|fflfflfflfflffl{zfflfflfflfflffl} |fflfflfflffl
ffl {zfflfflfflffl
ffl } |fflfflfflffl{zfflfflfflffl}
ously observed behavior of the wake behind a HAWT in boundary VI VII
V
layer flow [21]. The following sections will analyze the mechanisms
underlying the recovery of momentum and production of TKE in Given statistically stationary conditions, term I in Equation (2)
the wake. can be neglected, term III is ignored following the Boussinesq
approximation as the flow is relatively hydro-statically invariant,

Fig. 12. Crosswind profiles of TKE in the wake normalized by the incoming velocity at the midspan. Dotted lines denote the lateral edges of the VAWT rotor for reference.
 V.F.-C. Rolin, F. Porte-Agel / Renewable Energy 118 (2018) 1e13 7

Fig. 13. Vertical profiles of TKE in the wake normalized by the incoming wind-speed at the midspan. The dashed line represents the TKE in the incoming boundary layer flow.
Dotted lines denote the upper and lower edges of the VAWT rotor for reference.

and term V regarding molecular dissipation is omitted from this momentum, shown in Fig. 16, demonstrates that the high shear at
discussion as it was found to contribute little to the production of the periphery of the wake leads to turbulence-induced mixing of
MKE. Term IV could not be computed due to the lack of pressure momentum. Comparing the magnitude of momentum entrained
data collected in the present experiment. The streamwise gradients via turbulence and the mean flow shows that the mean flow plays
are not calculated due to the large distances between measurement the more significant role in reintroducing momentum into the
planes, though these gradients tend to have a limited influence on wake.
the production of TKE in comparison to the in-plane terms. Term VI, shown in Fig. 17, reflects the interaction of turbulence
Term II represents the quantity of mean kinetic energy advected and the mean flow. This term is predominantly negative, and
by the mean flow, shown in Fig. 14. overlaps with the regions of high TKE, indicating that energy me-
There is a region of positive advection in the upper region of the chanically produced as turbulence where the velocity gradient acts
wake, where the combined action of the vortices induces the highly against the Reynolds stresses is lost from the mean flow. Term VII,
energetic unperturbed flow above to moved down into the wake. shown in Fig. 18, illustrates another mechanism by which the
The advection of MKE is negative at the interface of the unper- interaction of turbulence and the mean flow produces MKE. This
turbed flow and the wake behind the negative-y sector of the rotor term achieves peak values where the gradient of the turbulence
as the wake is deflected into the unperturbed flow. The kinematic flux of momentum is highest and is positive where the velocity
advective flux of streamwise momentum, shown in Fig. 15, illus- decreases at a decreasing rate.
trates the impact of the counter-rotating vortex pair on trans- The MKE budget analysis reveals two phenomena that play
porting momentum into the wake. important roles in re-energizing the wake. The mean flow induced
Turbulence also tends to play a large role in mixing momentum by the counter rotating vortex pairs transports momentum into the
and re-energizing the wake. The total turbulence flux of streamwise wake from the faster moving flow and advects wake into the

Fig. 14. Advection of MKE at different downstream positions normalized by the rotor diameter and incoming flow at the midspan height. In-plane velocity vectors are overlaid. The
outline of the rotor is delineated by a black box.
8 V.F.-C. Rolin, F. Porte-Agel / Renewable Energy 118 (2018) 1e13

Fig. 15. Advective flux of streamwise momentum normalized by the incoming wind speed at the midspan height. In-plane velocity vectors are overlaid. The outline of the rotor is
delineated by a black box.

unperturbed flow. Turbulence plays a less straightforward role in

the MKE budget. The mechanical energy produced as turbulence is  
ve ve g 0 vU
lost from the mean flow where there is strong shear. The turbulent þ Uj ¼ þ di3 u0i qv
u0i u0j i
vt vx
|{z} |fflffl{zfflffl}j qv vxj
fluctuations on the edges of the wake also cause momentum to be |fflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflffl} |fflfflfflffl{zfflfflfflffl}
I III
entrained into the wake. Comparing the mean and turbulence II IV
  (3)
fluxes of momentum demonstrates that the mean flow plays the
v u0j e
larger role. 1 u0i p0



ε ;
vxj r vxj
|fflfflfflffl{zfflfflfflffl} |fflfflffl{zfflfflffl} |{z}
VII
V VI
3.2. Turbulence kinetic energy budget
where e represents TKE.
As previously discussed the highest levels of TKE observed in the Some terms of Equation (3) can be neglected given certain as-
wake of the VAWT coincide with the strongest velocity gradients on sumptions regarding the flow. Terms I and III of the budget equa-
the boundaries of the wake. In order to identify the mechanism tion yield negligible contributions to the production of TKE as the
responsible for producing turbulence we turn to the TKE budget wake is considered both statistically stationary and isothermal.
equation, given below, Term VI which is related to pressure oscillations could not be

Fig. 16. Turbulence flux of streamwise momentum normalized by the incoming wind speed at the midspan height. In-plane velocity vectors are overlaid. The outline of the rotor is
delineated by a black box.
 V.F.-C. Rolin, F. Porte-Agel / Renewable Energy 118 (2018) 1e13 9

Fig. 17. Production of MKE via turbulence at different downstream normalized by the rotor diameter and incoming flow at the midspan height with contours of velocity deficit and
in-plane velocity vectors overlaid. The outline of the rotor is delineated by a black box.

calculated due to the lack of pressure measurements, and the rate The amount of turbulence transported across the wake by the
of turbulence dissipation represented in Term VII could not be motion induced by the counter-rotating vortex pairs is of secondary
quantified due a similar lack of relevant data. The contribution of importance. This behavior is consistent with the production of
turbulent transport of TKE is found to be negligible in comparison turbulence in shear layers such as those in the wakes of HAWTs.
with that of the other terms, and is omitted from this section. As
mentioned earlier streamwise gradients are not presented due to 3.3. Counter-rotating vortex pairs
the large distances between measurement planes.
The remaining terms include the in-plane contributions to TKE As stated in the previous section, the wake of the VAWT features
production due to advection, and mechanical production (terms II two pairs of counter rotating vortices, clearly visible in Fig. 10. Via
and IV, respectively.) One should note that the production term is the analysis of the MKE budget, these vortices are shown to be the
actually the opposite of term VI in Equation (2), such that the principal cause of wake re-energization. It is therefore critically
shearing which detracted form the MKE budget contributes to the important to understand the cause of these vortices and under
production of TKE. The last remaining term to consider is that what circumstances will their effect be augmented or diminished.
related to the advection of turbulence, the magnitude of which is Ryan et al. [15] observed a similar CVP in the wake of a VAWT at low
presented in Fig. 19. Reynolds numbers using magnetic resonance velocimetry and
By comparison, it is clear that TKE is predominantly generated identify them as tip vortices caused by the lift-producing pressure
via shear between the wake and the unperturbed flow. differential over the rotor blades. Hofemann et al. [7] followed the

Fig. 18. Production of MKE via turbulence at different downstream normalized by the rotor diameter and incoming flow at the midspan height. In-plane velocity vectors are
overlaid. The outline of the rotor is delineated by a black box.
10 V.F.-C. Rolin, F. Porte-Agel / Renewable Energy 118 (2018) 1e13

Fig. 19. Advection of TKE at different downstream positions normalized by the rotor diameter and incoming flow at the midspan height. In-plane velocity vectors are overlaid. The
outline of the rotor is delineated by a black box.

trajectory of rotor tip vortices using PIV techniques in the rotor [24], divides the VAWT rotor into a series of parallel streamtubes
near-wake, and Tescione et al. [12] observed their presence and aligned with the incoming flow in order to more accurately capture
influence on the flow up to a distance of three diameters from the the changing aerodynamic behavior of the blades as they turn. Each
turbine, at which point these vortices tend to roll-up and dissipate. streamtube features two actuators for the two regions where the
Studies of HAWTs by Chamorro and Porte -Agel [22] demonstrate blades' trajectory crosses the streamtube. In this way the down-
that boundary layer turbulence causes rotor-tip vortices to break wind sector of the rotor will interact with the wake of the upwind
down earlier than in smooth flow. Given this information and the sector. Given the velocity relative the blades, the values of lift and
fact that previously observed rotor tip vortices are much smaller drag can be determined based on tabulated data found in the work
than the CVP in the present experiment and those reported by Ryan of Sheldahl and Klimas [25] and then modified to account for dy-
et al. [15], it is unlikely that these counter-rotating pairs are in fact namic stall using the MIT model, as implemented by Shamsoddin
tip vortices. and Porte -Agel [26].
Bastankhah and Porte -Agel [18] observed a similar CVP in the The flow relative to a blade in motion and the resulting forces
wake of yawed HAWTs in wind-tunnel experiments. They noted are illustrated in Fig. 20.
that this phenomenon is similar to the CVP associated with cross- !
It is clear from the figure that the force on the blade, F , is not
flow jets, a scenario in which a jet is oriented orthogonally to a !
necessarily aligned with the incoming flow, U ∞ . The model is
uniform flow. A literature review of crossflow jets can be found in
modified in order to incorporate the crosswind component of the
Mahesh [23], though it must be noted that the literature offers
lift and drag forces which previously were neglected. This is ach-
many possible reasons for why crossflow gives rise to a CVP without
ieved by yawing the actuators in each streamtube by an angle
proposing a definitive explanation. Bastankhah and Porte -Agel ! !
posited that counter-rotating vortex pairs occur in the presence of a g ¼ arctan F ,y= F ,x. A diagram of the model is provided in Fig. 21.
varying distribution of lateral velocity as a necessary result of The actuator induces a velocity aU∞ in the same direction as the
continuity. A more complete explanation of this mechanism can be force vector, where a is an induction factor. The value of the in-
found in the original work. Briefly, in the case of HAWTs, intro- duction factor is related to the thrust coefficient of the actuator by
ducing a yaw angle results in a crosswind force being exerted on the CT ¼ 4aðcos g
aÞ, and the thrust coefficient CT is equivalent to the
flow which in turn induces lateral velocity by conservation of total aerodynamic force exerted by an actuator normalized by the
momentum. Due to continuity, the crosswind gradient of lateral
velocity induces a vertical gradient of vertical velocity; these gra-
dients describe the crosswind vortices observed in the flow. Due to
the rotation of the VAWT, the blades exert both a streamwise and a
lateral force on the wind. As with yawed HAWTs, this crosswind
force introduces crosswind momentum to the flow, and conse-
quently gives rise to the circumstances we have associated with the
production of a CVP.

3.4. Analytical model

In order to investigate the underlying aerodynamics which

cause the VAWT rotor to exert a lateral force on the flow, an
analytical model is developed based on a blade-element-
momentum techniques adapted for VAWTs and yawed actuators.
The Double-Multiple Streamtube Model, described by Paraschivoiu Fig. 20. The flow relative to the blade and the resulting forces.
 V.F.-C. Rolin, F. Porte-Agel / Renewable Energy 118 (2018) 1e13 11

incoming wind speed and the area of the streamtube. The calcu-
lation is iterated, incorporating the induced flow into the velocity
relative to the turbine blade to recalculate the lift and drag. A more
detailed explanation of the model can be found in the Appendix.
By including the lateral component of the aerodynamic forces a
prediction can be made regarding the direction and magnitude of
the crosswind momentum introduced into the flow. This 2D anal-
ysis of the VAWT produces the distribution of velocity induced by
the rotor shown in Fig. 22.
The induced velocity predicted by the model is consistent with
the conditions which produce CVPs based on the previously pre-
sented continuity argument. This is an important result, as it
demonstrates that VAWTs induce forces on the flow in a manner
akin to yawed HAWTs. The two turbines therefore produce wakes
with similar characteristics, such as CVP and the deflection of the
momentum deficit from the centerline. It remains to be seen how
different VAWT design parameters and operating conditions will
alter the similarity between VAWTs and yawed HAWTs.
Fig. 22. Overhead view of the 2D VAWT rotor analyzed using the yaw-adapted Double-
Multiple Streamtube model with the direction and magnitude of the velocity induced
4. Summary
by the rotor illustrated by the blue vectors. The direction and magnitude of the
incoming flow is denoted by the red arrow. The VAWT's direction of rotation is illus-
Using stereo-PIV techniques in a wind tunnel, the wake behind a trated by the circular black arrow. (For interpretation of the references to colour in this
small VAWT immersed in a boundary layer is quantified and figure legend, the reader is referred to the web version of this article.)
analyzed. Several interesting characteristics of the wake are
observed. Possibly the most notable feature of the flow is the
presence of two pairs of counter-rotating vortices at the edges of Due to the boundary layer flow, the wake recovers asymmetri-
the wake. Analyzing the MKE budget and fluxes of momentum in cally as more momentum is entrained from the faster moving flow
the wake indicates that these CVPs are the most important mech- above the wake by the counter-rotating vortices. The high velocity
anism in introducing momentum to the wake. These counter- gradient on the boundary of the wake behind the negative-y sector
rotating vortex pairs tend to occur where there is a variation in of the rotor gives rise to peak turbulence levels and encourages
crossflow momentum and have also been observed in the wakes of mixing in this region. Through the previously discussed analysis of
yawed HAWTs due to the lateral force these rotors produce. An the MKE budget, it appears that the turbulence flux of streamwise
analytical model is developed based on the Double-Multiple momentum does not play as important a role as the mean flux in
Streamtube model commonly applied to VAWTs and is adapted to terms of re-energizing the wake. The boundary layer also causes the
incorporate yawed actuators. The model demonstrates that VAWTs turbulence added to the flow to be greater at the upper boundary
are expected to exert lateral forces and induce flow consistent with between the wake and the unperturbed flow than at the lower
the production of CVPs observed in the wake. VAWT wakes appear boundary due to the higher shear in this region as a result of the
to have much in common with the wakes of yawed HAWTs, faster moving flow higher in the boundary layer.
including not only the production of CVPs but also the lateral It remains unclear why there is such a discrepancy between
deflection of the wake center in the same direction as the velocity VAWT wakes observed in literature. In the case studied here it
induced by the rotor. seems clear the turbulence is primarily produced via shear, and that

Fig. 21. Schematic illustration of the Double Multiple Streamtube model with yawed actuators.
12 V.F.-C. Rolin, F. Porte-Agel / Renewable Energy 118 (2018) 1e13

the stronger the velocity deficit, the higher the turbulence. It may
prove to be the case that the low tip-speed ratio implemented in ! !
W $nb
this study results in highly asymmetric aerodynamic behavior, as a ¼ arctan ! : (A.6)
the flow relative to the blade will fluctuate highly as the VAWT W $b
c
rotor spins. The analytical model developed here, once validated,
The coefficients of lift and drag, Cl and Cd respectively, are
may prove useful to predicting how tip-speed ratio is related to
extrapolated from data provided for NACA0018 airfoils by Kumar
these oscillatory aerodynamics and wake asymmetry.
et al. [27] and Sheldahl and Klimas [25]. Dynamic stall effects are
reproduced via the modified MIT model developed by Noll and
Ham [28] which is a modification of the original MIT model [29].
Acknowledgments Shamsoddin and Porte -Agel [30] previously implemented this
model in an LES study of VAWT wakes and provided a thorough
This research was supported by the Swiss National Science explanation of how to do so for others to follow. The drag force is
Foundation [grant 200021_172538, and grant 206021_144976], the aligned with the flow relative to the blade and therefore has a unit
Swiss Federal Office of Energy [grant SI/501337-01], and the Swiss vector
Innovation and Technology Committee (CTI) within the context of
the Swiss Competence Center for Energy Research ’FURIES: Future !
Swiss Electrical Infrastructure’. The Authors would also like to b W
d¼ : (A.7)
thank Majid Bastankhah for his help setting up and performing PIV. jjWjj
The unit vector of lift is normal to the drag vector,

Appendix A
!
bl ¼ W ⊥ : (A.8)
We examine the aerodynamics of vertical-axis wind-turbines in jjWjj
order to investigate the lateral forces produced by the rotor. To this
purpose we apply a modified version of the Double-Multiple The total force on the blades is
Streamtube Model described by Paraschivoiu [24]. This model
segments the rotor into a series of parallel streamtubes aligned ! 1  2  
T ¼ rcHBW  Clbl þ Cd b
d ; (A.9)
with the incoming flow. Each streamtube features two actuators, 2
which represent the two zones where the blades intersect the flow.
The present study suggests modifying the actuators to account for where B, c, H are respectively the number of blades, the blade chord
the influence of lateral forces. length, and the blade span. Reciprocally, the thrust force exerted on
!
The incoming velocity is defined as the streamtube is
T . This vector consists of both the streamwise
and lateral components of thrust. The yaw angle then follows as the
!
U ¼ U∞bi; (A.1) argument of the thrust vector

where U∞ is the magnitude of the incoming flow, and bi is the unit ! !

T ,bj
vector aligned with the streamwise flow. We consider an actuator g ¼ arctan ! : (A.10)
which crosses the streamtube with a yaw angle g. During the up-
T ,bi
wind portion of the cycle, where the incoming flow is unperturbed, For a streamtube encapsulating the sector of the rotor between
we assume an induction factor a which induces a velocity normal to q
Dq and q þ Dq, the overall actuator thrust is the mean thrust
the actuator, over the streamtube as a portion of the number of streamtubes
!  
U a ¼
aU∞ cos gbi þ sin gbj (A.2) qZþDq
1 1  2 qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
T¼ rcHB W  C 2 þ C 2 dq: (A.11)
We assign a coordinate system to the blade's frame of reference 2 2NDq l d
such that the unit vector aligned with the chord line is q
Dq

By conservation of momentum, the trust exerted by the actuator

c ¼ sin qbi
cos qbj;
b (A.3)
should be equivalent to the product of the velocity normal to the
actuator and the total change in mass-flow-rate in the streamtube,
where q is the azimuthal angle between the blade's position and
the abscissa, and bj is the unit vector perpendicular to the incoming
flow in the absolute frame of reference. The unit vector normal to T ¼ rAs ðU∞ cos g
aU∞ ÞðU∞
ð1
2aÞU∞ Þ; (A.12)
the chord is therefore
where As is the streamtube area. For streamtubes aligned with the
b ¼ cos qbi þ sin qbj:
n (A.4) x-axis, the upwind azimuthal angle ranges between p2 and 32p, the
frontal area of a streamtube centered around azimuthal angle q is
The tangential velocity of a blade is, by definition, lU∞ , where l
is the tip-speed-ratio, or the ratio of the blade tangential velocity  
 q Dq
tot the incoming flow. The total relative velocity with respect to an As ðqÞ ¼ DHcos sin ; (A.13)
2 2
individual blade is therefore

! ! ! where D is the rotor diameter. Equating this expression to the

W ¼ U þ U a þ lU∞ b
c: (A.5) magnitude of the thrust in Equation (A.11) and normalizing by
1 rA U 2 yields the following relationship for the induction factor
The angle of attack between the flow relative to the blade and 2 s ∞
the blade's direction of motion is and thrust for a single streamtube
 V.F.-C. Rolin, F. Porte-Agel / Renewable Energy 118 (2018) 1e13 13

[9] L. Battisti, L. Zanne, S. Dell'Anna, V. Dossena, G. Persico, B. Paradiso, Aero-

Z qþDq   qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
jjWjj 2 dynamic measurements on a vertical Axis wind turbine in a large scale wind
s C 2l þ C 2d dq tunnel, J. Energy Resour. Technol. 133 (3) (2011) 031201.
q
Dq U∞
CT ¼ 4aðcos g
aÞ ¼   ; (A.14) [10] J. Bossard, Characterisation experimentale du decrochage dynamique dans les
  hydroliennes a flux transverse par la methode PIV (Particle Image Velocim-
2N Dqcos 2q sin Dq2 etry). Comparaison avec les resultats issus des simulations numeriques, Uni-
versite de Grenoble, 2013. Ph.D. thesis.
[11] L. Beaudet, Etude expe rimentale et nume rique du de crochage dynamique sur
where s is the rotor solidity, defined as Bc D . For values of thrust une eolienne a  axe vertical de forte solidite, Universite de Poitiers, 2014. Ph.D.
CT > 0:96, Glauert's empirical correction is applied. thesis.
[12] G. Tescione, D. Ragni, C. He, C.J.S. Ferreira, G.J.W. van Bussel, Near wake flow
Equations A.2eA.14 can be iterated until convergence. This
analysis of a vertical axis wind turbine by stereoscopic particle image veloc-
process is repeated for the downwind portion of the rotor, using imetry, Renew. Energy 70 (2014) 47e61.

! h i [13] P. Bachant, M. Wosnik, Characterising the near-wake of a cross-flow turbine,

U þ 2aU∞
cos gbi
sin gbj (A.15) J. Turbul. 16 (4) (2015) 392e410.
-Agel, Wind-tunnel study of the wake behind a vertical axis
[14] V. Rolin, F. Porte
wind turbine in a boundary layer flow using stereoscopic particle image
as the incoming streamtube velocity. The magnitude of this velocimetry, J. Phys. Conf. Ser. 625 (2015) 012012. ISSN 1742-6588, http://
stacks.iop.org/1742-6596/625/i¼1/a¼012012?key¼crossref.
incoming velocity should also be used when normalizing values 276a4e4067e30f10c25e95f0c50e7beb.
with respect to the downwind actuators. This procedure can be [15] K.J. Ryan, F. Coletti, C.J. Elkins, J.O. Dabiri, J.K. Eaton, Three-dimensional flow
used to predict the direction of the velocity induced by the VAWT. field around and downstream of a subscale model rotating vertical axis wind
turbine, Exp. Fluids 57 (3) (2016) 1e15.
This model has some shortcomings and limitations such as the [16] M. Khosravi, P. Sarkar, H. Hu, An experimental investigation on the wake
fact that it does not account for the deflection or expansion of the characteristics of a wind turbine in an atmospheric boundary layer wind, AIAA
streamtubes nor the influence of one streamtube on another. 1732 (2016). ISSN 08899746.
[17] A. Posa, C.M. Parker, M.C. Leftwich, E. Balaras, Wake structure of a single
Nonetheless, due to the fluctuating orientation of the blades with vertical axis wind turbine, Int. J. Heat Fluid Flow 61 (Part A) (2016) 75e84.
the incoming wind, relatively large lateral forces can be incurred [18] M. Bastankhah, F. Porte -Agel, Experimental and theoretical study of wind
and cannot be neglected as they will influence the locally induced turbine wakes in yawed conditions, J. Fluid Mech. 806 (2016) 506e541.
[19] P. Bachant, M. Wosnik, Effects of Reynolds number on the energy conversion
velocity. This model takes steps to incorporate this lateral force into and near-wake dynamics of a high solidity vertical-axis cross-flow turbine,
the momentum balance for the VAWT and improve its robustness Energies 9 (2) (2016) 1e18.
as a means of evaluating the performance of a turbine. [20] D.B. Araya, J.O. Dabiri, A comparison of wake measurements in motor-driven
and flow-driven turbine experiments, Exp. Fluids 56 (2015) 150.
[21] L. Vermeer, J. Srensen, A. Crespo, Wind turbine wake aerodynamics, Prog.
References Aerosp. Sci. 39 (6e7) (2003) 467e510.
[22] L.P. Chamorro, F. Porte -Agel, A wind-tunnel investigation of wind-turbine
[1] J. O. Dabiri, Potential order-of-magnitude enhancement of wind farm power wakes: boundary-layer turbulence effects, Boundary-Layer Meteorol. 132
density via counter-rotating vertical-axis wind turbine arrays, J. Renew. Sus- (1) (2009) 129e149.
tain. Energy 3(4). [23] K. Mahesh, The interaction of jets with crossflow, Annu. Rev. Fluid Mech. 45
n, L. Castillo, H.S. Kang, C. Meneveau, Experimental study of
[2] R.B. Cal, J. Lebro (2013) 379e407. First.
the horizontally averaged flow structure in a model wind-turbine array [24] I. Paraschivoiu, Wind Turbine Design with Emphasis on Darrieus Concept,
boundary layer, J. Renew. Sustain. Energy 2 (2010) 1e25. ISSN 19417012. Presses internationales Polytechnique, Quebec, 2002. ISBN 9782553009310.
[3] R.J. Muraca, R.J. Guillotte, Wind Tunnel Investigation of a 14ft Veritcal axis [25] R.E. Sheldahl, P.C. Klimas, Aerodynamic Characteristics of Seven Symmetrical
Windmill, NASA Langley Research Center, Hampton, Virginia, 1976. Tech. Rep. Airfoil Sections through 180-degree Angle of Attack for Use in Aerodynamic
[4] G. Brochier, P. Fraunie, C. Beguier, I. Paraschivoiu, Water channel experiments Analysis of Vertical axis Wind Turbines, Sandia National Laboratories, Albe-
of dynamic stall on Darrieus wind turbine blades, J. Propuls. 2 (5) (1986) querque, 1981. Tech. Rep.
445e449. [26] S. Shamsoddin, F. Porte -Agel, Large eddy simulation of vertical axis wind
[5] G. Bergeles, A. Michos, N. Athanassiadis, Velocity vector and turbulence in the turbine wakes, Energies 7 (2) (2014) 890e912. ISSN 19961073.
symmetry plane of a Darrieus wind generator, J. Wind Eng. Ind. Aerod. 37 (1) [27] V. Kumar, M. Paraschivoiu, I. Paraschivoiu, Low Reynolds number vertical axis
(1991) 87e101. wind turbine for Mars, Wind Eng. 34 (2010) 461e476.
[6] N. Fujisawa, S. Shibuya, Observations of dynamic stall on Darrieus wind tur- [28] R.B. Noll, N.D. Ham, Dynamic Stall of Small Wind Systems, Aerospace Systems
bine blades, J. Wind Eng. Ind. Aerod. 89 (2) (2001) 201e214. Inc, Burlington, MA, USA, 1983. Tech. Rep.
[7] C. Hofemann, C.J.S. Ferreira, K.R. Dixon, G.J.W. van Bussel, G. a. M. van Kuik, [29] N.D. Ham, Aerodynamic loading on a two-dimensional airfoil during dynamic
A. Maria, F. Scarano, G.V. Bussel, G.V. Kuik, 3D stereo PIV study of tip vortex stall, AIAA J. (1968) 1927e1934.
evolution on a VAWT, in: Wind Energy Conference and Exhibition EWEC, [30] S. Shamsoddin, F. Porte -Agel, A large-eddy simulation study of vertical axis
European Wind Energy Association EWEA, Brussels, 2008, pp. 1e8. wind turbine wakes in the atmospheric boundary layer, Energies 9 (5) (2016)
[8] C.J.S. Ferreira, The Near Wake of the VAWT 2D and 3D Views of the VAWT 1e23. ISSN 19961073.
Aerodynamics, Universidade Tecnica de Lisboa, 2009. Ph.D. thesis.