Energy for Sustainable Development 45 (2018) 142–149

Contents lists available at ScienceDirect

Energy for Sustainable Development

An approach for the optimization of diffuser-augmented hydrokinetic

blades free of cavitation
Déborah A.T.D. do Rio Vaz a , Jerson R.P. Vaz a, * , Paulo A.S.F. Silva b,1
a
Natural Resources Engineering Graduate Program, Institute of Technology, Federal University of Pará, Av. Augusto Correa, N 1 Belém, PA 66075-900, Brazil
b
University of Brasília, Faculty of Mechanical Engineering, Campus Darcy Ribeiro, Brasília, DF 70910-900, Brazil

A R T I C L E I N F O A B S T R A C T

Article history: Due to the Venturi effect caused by a diffuser, which speed-up the velocity through the rotor, shrouded tur-
Received 16 December 2017 bines are able to exceed the Betz-Joukowsky limit if the power coefficient is based on the rotor diameter.
Received in revised form 19 May 2018 However, on hydrokinetic turbines this increased velocity may also promote cavitation on the blade. As this
Accepted 6 June 2018 subject is still not clear on the current literature, this work presents a novel approach for optimizing hydroki-
Available online 21 June 2018
netic turbines free of cavitation under diffuser effect. The model uses the minimum pressure coefficient as
the criterion to keep the pressure near blade tip above water vapor pressure. It includes an extension of Vaz
Keywords: & Wood’s optimization in order to take into account the influence of the diffuser speed-up ratio regarding
Cavitation
cavitation effect. A changing on the thrust coefficient is assumed to optimize chord and twist angle distri-
Blade optimization
butions along the blade. As a result, the proposed model shows that cavitation is indeed sensitive to the
Hydrokinetic turbines
Diffuser diffuser speed-up ratio, demonstrating that such a phenomenon needs to be considered in the design of
BET diffuser-augmented hydrokinetic turbines. Also, the optimization method corrects the chord without rele-
vant changing in the turbine power coefficient, where the increased power output is about 42% higher than
the bare turbine for a water velocity of 2.5 m/s. In this case, the model is assessed through comparisons
using a 3-bladed hydrokinetic turbine with 10 m diameter, in which the diffuser speed-up ratio is varied.
Furthermore, an evaluation is made with models available in the literature, suggesting good performance
concerning the cavitation analysis on shrouded rotor design with the proposed optimization procedure.
© 2018 International Energy Initiative. Published by Elsevier Inc. All rights reserved.

Introduction
It is well known that the diffuser technology can increase the
axial velocity through the rotor, improving the power output (Chen,
Ponta, & Lago, 2011; Vaz & Wood, 2018). It means that a compact sys-
tem can convert a greater amount of energy if the power coefficient
is based on the rotor diameter (Sørensen, 2016). This advance has
gained attention over the last decades as an alternative to conven-
tional energy systems, especially on locations constrained by shallow
waters or low stream velocity (Anyi & Kirke, 2010). As a consequence,
some works in the current literature have reported the use of diffuser
on hydrokinetic or tidal turbines (Belloni, Willden, & Houlsby, 2017;
Dominguez, Achard, Zanette, & Corre, 2016; Silva et al., 2017). Among
them, only a few takes into account the cavitation in order to design
hydro blades. For example, recently Silva et al. (2017) developed an
* Corresponding author.
E-mail address: jerson@ufpa.br (J.R.P. Vaz).
1
Currently at Centre for Computational Engineering Sciences, Cranfield University,
Cranfield MK43 0AL, United Kingdom.
optimization approach based on Blade Element Theory (BET) for hor-
izontal axis hydrokinetic turbines considering the possibility of cavi-
tation, where the minimum pressure coefficient is the criterion used
for identifying cavitation on blades. Their model has demonstrated
good behavior, and indeed minimizes cavitation inception. But, it is
applied only for bare hydrokinetic turbines. Batten, Bahaj, Molland,
and Chaplin (2008) developed a BET model for hydrodynamic design
of marine current turbines. In their model, the prediction of cav-
itation is carried out for certain cases with relatively shallow tip
immersion. It is found that cavitation could be avoided with the
use of suitable designs and choice of 2D sections. Even though their
work presented very interesting thoughts on cavitation, studies on
optimization of hydrokinetic blades free of cavitation are not done.
There is an extensive literature on diffuser-augmented turbines.
But, most of them are related to wind rotors. Jafari and Kosasih
(2014) performed a Computational Fluid Dynamics (CFD) study, in
which the augmentation is heavily dependent on the diffuser geom-
etry. They reported that a higher area ratio creates greater pressure
reduction at the diffuser outlet, which increases the mass flow rate.
Abe and Ohya (2004) combined a numerical and experimental inves-
tigation of a shrouded turbine with a flanged diffuser technology.

https://doi.org/10.1016/j.esd.2018.06.002
0973-0826/ © 2018 International Energy Initiative. Published by Elsevier Inc. All rights reserved.
 D. do Rio Vaz et al. / Energy for Sustainable Development 45 (2018) 142–149 143

They found that the turbine thrust coefficient using diffuser is
smaller than for a bare wind turbine. Ohya and Karasudani (2010)
developed a turbine within a diffuser shroud with a broad-ring brim
at the exit. The shrouded wind turbine power is increased by a fac-
tor between 2 and 5 over a bare wind turbine in the same operating
condition. It shows the importance of the development of models
able to optimize shrouded hydrokinetic blades free of cavitation.
Rio Vaz, Mesquita, Vaz, Blanco, and Pinho (2014) developed impor-
tant formulations for the performance analysis of wind turbine with
diffuser based on BET. Their results yielded good agreement with
experimental data. Based in their work, Vaz and Wood (2016) stated
the equations for the aerodynamic optimization of a wind turbine
with diffuser. In this case, it is assumed that the same conditions for
the axial velocity in the wake of an ordinary wind turbine can be
applied on the flow far downwind of the diffuser outlet, as proposed
by van Bussel (1999).
Therefore, as the authors are unaware of any study of blade
optimization free of cavitation under diffuser effect, an optimiza-
tion procedure for shrouded hydrokinetic blades is proposed in this
work. The methodology is an extension of Vaz and Wood (2016)
optimization, where the minimum pressure coefficient criterion is
incorporated into the turbine thrust coefficient in order to keep the
pressure near blade tip above water vapor pressure. The chord of the
blade is determined using a one-dimensional analysis, where a cor-
rection to avoid cavitation is employed. A hydrokinetic turbine with
10 m diameter is used to evaluate the effect of the diffuser speed-
up ratio concerning the cavitation. To assess the proposed approach,
comparisons with other models available in the literature are made.
As a result, the model shows that cavitation is sensitive to the dif-
fuser speed-up ratio, suggesting that such a phenomenon needs to
be considered in the design of shrouded hydrokinetic turbines. In
addition, the optimization method corrects the chord without rele-
vant changing in the turbine power coefficient. Also, the shrouded
turbine power output increases about 42% when compared with the
bare turbine for a water velocity of 2.5 m/s.
The remainder of this paper is organized as follows. The next
section shows the Cavitation criterion used in the model. Section
3 presents the mathematical modeling for the Optimum design of
diffuser-augmented hydrokinetic blades, in which a simple one-
dimensional axial momentum theory with a diffuser is carried out.
This section also shows the expressions for the changing in the thrust
coefficient, which leads to a corrected formulation of the chord dis-
tribution free of cavitation. In Section 4, the Results and discussion
are stated, where the performance and comparisons of the proposed
model is presented. Section 5 shows the Conclusions of this study.
q is the density, p and patm are the local and atmospheric pressures
respectively, and the relative velocity of water on each blade section
is defined by Vaz and Wood (2016) as

2 2
W= [cV0 (1 − a)] + [Yr (1 + a )] . (3)
The parameter c is the diffuser speed-up ratio, and it is responsible
for modifying the axial velocity on the rotor. The free-stream velocity
corresponds to V0 . The parameters a and a are the axial and tan-
gential induction factors, respectively, while Y and r are the angular
velocity and radial position of the turbine. The formulation for the
number of cavitation s is (Shinomiya, 2015)
patm + qgH − pv
s= , (4)
1 2
2 qW
where g is the gravitational acceleration, H is the submerged distance
and pv is the vapor pressure. Substituting Eq. (4) in Eq. (1), it gets
VCAV ≥ W, where the cavitation velocity is

patm + qgH − pv
VCAV = . (5)
− 12 qCp min
Note that the cavitation can be avoided whether the relative veloc-
ity on each blade section is lower than the local cavitation velocity.
This condition is fundamental for the optimization methodology
described below in which the chord length is corrected at each
section when VCAV < W. Fig. 1 illustrates the static pressure condi-
tion on a DAHT blade section. In this case, the rotor is located near
the inlet of the diffuser.
Optimum design of diffuser-augmented hydrokinetic blades
The optimum expressions for shrouded turbines come from the
momentum equations with rotational velocities in the flow (Fletcher,
1981; Philipis, 2003; Sørensen, 2016). According to Rio Vaz et al.
(2014) for modern turbines it is necessary to consider the effect
of the tangential induction factor, a . The elemental torque can be
obtained directly from the momentum equation applied to the con-
trol surface shown in Fig. 2, in which the infinitesimal area at the

Cavitation criterion

According to Adhikari, Vaz, and Wood (2016), cavitation effect

typically occurs in hydro turbines, usually leading to vibration, blade
surface damage and performance loss. Therefore, those issues need
to be avoided in hydro rotors. Hence, as the diffuser increases the
flow axial velocity at the rotor plane, it is necessary to include a
restriction to avoid cavitation in the design of hydrokinetic blades.
The most used criterion to minimize or avoid cavitation on hydro
turbines relates the number of cavitation (s) with the minimum
pressure coefficient (Cp min ) through the expression

Cp min + s ≥ 0, (1)

where Cp min is the minimum value of the pressure coefficient Cp ,

defined by

p − patm
Cp = 1
, (2)
2
2 qW Fig. 1. Simplified illustration of the static pressure condition on a DAHT blade section.
144 D. do Rio Vaz et al. / Energy for Sustainable Development 45 (2018) 142–149

Fig. 2. Simplified illustration of the flow velocities through an ideal DAHT.

rotor plane is dA = 2prdr, allowing the power coefficient as (Rio Vaz

et al., 2014; Vaz & Wood, 2016)
Fig. 3. Velocity diagram for the section of the rotor blade (Vaz & Wood, 2016).
k
8
CP = ca (1 − a)x3 dx, (6)
k2
0 Also, as described by Silva et al. (2017), cu can be given as

where x = Yr/V0 and k = YR/V0 are the local-speed ratio and the  2
tip-speed ratio, respectively. Eq. (6) will be used to optimize the axial 2pr CT V0
cu = . (10)
induction factor, a, later. B Cn W
So, the main contribution of this work is to combine the opti-
mization approaches recently developed by Silva et al. (2017) and Therefore, to calculate the corrected chord distribution, it is nec-
Vaz and Wood (2016). The former presented a good formulation to essary to replace W by (1 − fS )VCAV in Eq. (10), yielding
avoid cavitation on hydrokinetic blades, however it is applied only
for non-shrouded hydrokinetic turbines. The latter demonstrated a  2
2pr CT V0
very interesting blade optimization equations, but it works only for cc = . (11)
B Cn (1 − fS )VCAV
shrouded wind rotors, obviously without cavitation prediction. Thus,
the diffuser contribution is incorporated into BET by using the trian-
Combining Eqs. (10) and (11), it gets:
gle diagram shown in Fig. 3. The flow velocity at the blade section
is given by c(1 − a)V0 . The elemental normal and tangential force  2
coefficients at any location on a blade are defined as in BET for a bare W
cc = cu , (12)
turbine. Therefore, the optimization procedure proposed here for the (1 − fS )VCAV
chord distribution, cu , is made similarly as in Vaz and Wood (2016),
for which where fS is an arbitrary safety factor defined in the interval 0 ≤ fS < 1
over the blade element for the calculation of the chord length. This
8pra F sin 0 cos 0 safety factor is used here to introduce a margin of security against
cu = , (7)
(1 + a )BCt cavitation risks, in order to ensure (1 − fS )VCAV < W, leading to
cc > cu . Note that Eq. (12) is the same obtained by Silva et al. (2017).
where B is the number of blades, Ct = CL sin0 − CD cos0 is the tan- It is worth noting that the diffuser effect occurs only through c in
gential force coefficient, CL and CD are, respectively, the lift and drag Eqs. (8) and (9). The Prandtl’s tip loss factor F is used in Eq. (7), which
coefficients of the airfoil comprising the element. The flow angle, 0, is defined as the ratio of the total bound circulation of the blades and
is obtained from the velocity diagram shown in Fig. 3 as the circulation of a rotor with an infinite number of blades (Vaz &
Wood, 2016; Wald, 2006). BET calculations made with the Prandtl
(1 − a)V0 factor show good agreement with free wake vortex theory and test
tan 0 = c . (8)
(1 + a )Yr data according to Favacho et al. (2016) and Branlard (2013).
The optimum design for a hydrokinetic blade under diffuser effect
or
is obtained through maximization of the power coefficient using the
a Yr term ca (1 − a) of Eq. (6). This yields
tan 0 = (9)
caV0  
d da
[ca (1 − a)] = c (1 − a) − a = 0 (13)
if the local angles of attack are below stall. da da
 D. do Rio Vaz et al. / Energy for Sustainable Development 45 (2018) 142–149 145

Since c = 0, Eq. (13) can be simplified as: Table 1

Design parameters used in the present simulation (Shinomiya, 2015; Silva et al., 2017).

Parameters Values
da
(1 − a) = a . (14)
da Turbine diameter (D) 10.0 m
Hub diameter 1.5 m
Number of blades 3
As reported in Vaz and Wood (2016), combining Eqs. (8) and (9), Free-stream velocity (V0 ) 2.5 m/s
it gives Water density (q) at 25 ◦ C 997 kg/m3
Submergence of the turbine (H) 6m
Patm 1 × 105 Pa
x2 a (1 + a ) = c2 a(1 − a) (15) Pv 3.17 × 103 Pa
Gravity (g) 9.81 m/s2
Safety factor (fS ) 5%
Eq. (15) differentiated with respect to a yields Angular velocity (Y) 35 rpm

 2
da x
(1 + 2a ) = 1 − 2a. (16) Results and discussion
da c
In this section, the results of the proposed model are discussed
So, if Eqs. (14) and (16) are combined with Eq. (15), the optimum and analyzed. Design information and operating conditions of the
relationship between a and a becomes: hydrokinetic turbine are presented. For the model, a sensitivity anal-
ysis is performed on a specific section of the blade, showing the
results obtained for the optimization procedure with and without
a opt = (1 − 3aopt )/(4aopt − 1). (17) cavitation correction on the chord distribution in order to assess the
approach under diffuser effect. In addition, results using the cor-
Eq. (17) is similar to the obtained by Glauert and Durand (1963) rected model are evaluated regarding the insertion of the cavitation
for bare turbines. This occurs because in this approach the diffuser constraint varying radially.
effect is considered only on the axial direction through the parameter
c. Thus, substituting Eq. (17) in Eq. (15), the optimum relationship Operating condition of the hydrokinetic turbine
for the axial induction factor is
To analyze the performance of the model, it is considered the hor-
 2
 2
x x izontal axis hydrokinetic turbine designed and simulated by Silva
16a3opt − 24a2opt + 9−3 aopt + − 1 = 0. (18) et al. (2017), where the parameters of the turbine are described in
c c
Table 1. Those parameters have been considered due to the possibil-
ity of cavitation at the rotor tip. For the turbine blades, the hydrofoil
Eq. (18) is valid only if the local angles of attack are below stall, NACA 653 − 618 is used, and its hydrodynamic parameters such as
in which a and a are not independent since the force according to lift, drag, and minimum pressure coefficients are obtained by the free
potential flow theory is perpendicular to the local velocity seen by software XFOIL (Drela, 1989), which is a coupled panel/viscous code
the blade (Hansen, 2008; Wood, 2015). Also, it is important to note developed at MIT. XFOIL is a collection of programs for airfoil design
that the present optimization procedure is valid for k > 1 approx- and analysis for incompressible/compressible viscous flows over an
imately, and further details in this regard may be seen in Vaz and arbitrary airfoil. In this code, a zonal approach is used to solve the
Wood (2016) . viscous flow indirectly and an equivalent inviscid flow is postulated
The iterative algorithm for the calculation of optimum chord and outside a displacement streamline that includes the viscous layer,
twist angle, starting at its most external section, is described below, becoming a powerful software for hydro and aerodynamic design,
using the following as input: r, c, CL (a opt ), CD (a opt ) and V0 for a and presenting good agreement when compared with experimental
given k. data (Benini, 2004; Favacho et al., 2016).
146 D. do Rio Vaz et al. / Energy for Sustainable Development 45 (2018) 142–149

Sensibility of the model to the diffuser effect
In order to verify the sensibility of the model, the data in Table 2
is obtained using BET only for a section at the radial position
r/R = 0.82. The solidity is s s = 0.029, twist angle is b = 3.29◦ ,
and chord length with no correction is cu,opt /R = 0.05. Note from
the results shown in Table 2 that increasing the freestream velocity,
the angle of attack is increased, while the minimum pressure coef-
ficient decreases. In this case, the cavitation inception occurs for
a = 8.8◦ , from which Cp min + s < 0, as depicted in Fig. 4a. This
result is consistent with that found by Goundar, Ahmed, and Lee
(2012), which developed a numerical and experimental study on a
cavitation free hydrofoil applied to marine current turbine rotors.
They demonstrated that Cp min for a NACA 63815 near the blade
tip can heavily decrease with a between 10◦ and 24◦ . Fig. 4b is
generated using Eq. (12), which shows the correction on the chord
regarding cavitation. The ratio cc /cu is sensitive to the diffuser
speed-up ratio, c. For values of angle of attack above 8.8◦ , the cav-
itation increases, leading to an increased chord as a consequence.
This occurs because the optimization procedure modifies the chord
in order to avoid the cavitation inception. This may be seen through
the relative velocity which is sensitive to the diffuser effect, as ver-
ified in Eqs. (3) and (12). Another important result of the model
is the fact that the thrust coefficient, shown in Fig. 5, decreases
when the diffuser speed-up ratio increases remaining below to the
Betz-Joukowsky limit. This behavior strongly agree with the result
obtained by Hansen, Sorensen, and Flay (2000), which evaluated a
one-dimensional CFD analysis to study a diffuser-augmented wind
turbine with a NACA 0015 airfoil on an ideal turbine. It can be
noted that the mass flow through the turbine is affected by changes
of the thrust coefficient in the shrouded turbine. Consequently, the
increased flow velocity at the rotor plane, induced by the diffuser,
also lead to an increased power output for the same thrust coeffi-
cient compared to a bare turbine. For example, for V0 = 2.5 m/s,
the thrust coefficient decreases from 0.888 (c = 1.0 — bare
turbine) to 0.885 (c = 2.0 — shrouded turbine). Even though
Hansen et al. (2000) have studied shrouded wind turbine, lift-based,
axial-flow turbines use the same principles as air-craft wings, and
propellers, as further described by Laws and Epps (2016). Hence,
such principles are valid for hydro turbines as well.
Cavitation effect on the optimization method
To evaluate the cavitation effect on the present approach, the
design parameters in Table 1 are used. The results shown in Fig. 6
are obtained for H = {5.5, 7.5, 9.5} and c = {1.0, 1.5, 2.0}. During
the analysis, the blade needs to be located as in Fig. 1, where each
section is positioned in a depth h = H − r in relation to the
water free surface. Fig. 6a is generated without cavitation correc-
tion so the condition W > VCAV can be reached, whereas Fig. 6b is
performed using the correction through Eq. (12) when W < VCAV .
The submergence of the turbine, H, strongly affect the pressure con-
dition on the blade tip. Observe in Fig. 6a as much as deeper the
turbine rotor, the cavitation velocity, VCAV , increases and so the cav-
itation inception may be avoided, as demonstrated through Eq. (5).
Independent of the submergence H, the optimization model keeps
the relative velocity always below cavitation velocity (Fig. 6b). This
is a very important feature because the hydrokinetic rotor can be

Table 2
Parameters for a typical blade section.

V0 a - (◦ ) 0-(◦ ) W-(m/s) Rec Cp min s

2.00 0.7 3.99 15.14 4342450 −0.8305 1.0100

2.15 2.8 6.09 15.16 4348119 −0.8729 1.0074
2.30 4.8 8.09 15.19 4354439 −0.9068 1.0034
2.45 6.9 10.19 15.21 4361443 −0.9407 1.0008
2.60 8.8 12.09 15.24 4368816 −0.9746 0.9968
2.75 10.6 13.89 15.26 4376554 −1.0085 0.9942
2.90 12.4 15.69 15.29 4384667 −1.1102 0.9903
3.05 14.0 17.29 15.32 4393209 −1.3475 0.9864
3.20 15.7 18.99 15.35 4402210 −1.5932 0.9826
3.35 17.4 20.69 15.38 4411580 −1.8220 0.9788

Fig. 4. Effect of the minimum pressure coefficient to: (a) the number of cavitation (for c = 1), and (b) the chord ratio.
 D. do Rio Vaz et al. / Energy for Sustainable Development 45 (2018) 142–149 147

When the speed-up ratio, c, is higher than 1, the cavitation

Fig. 5. Thrust coefficient as a function of the freestream velocity.
designed free of cavitation for any H. On the blade shape, the major
consequence is the increasing of the chord distribution, as depicted
in Fig. 7a.
inception occurs early, and then a correction is needed. Fig. 7a shows
a comparison between the present model, and others available in
the literature, including the classical Glauert’s optimization (Glauert
& Durand, 1963). The model starts the cavitation correction of the
chord distribution at approximately 60% of the rotor span when
c = 1.3. For a radial position of 90% of the blade, the cavitation
correction leads to an increased chord length of 64% compared to
Vaz and Wood (2016) model. It is worth noting that the diffuser
changes the chord length, intensifying the possibility of cavitation
due to the increasing of the axial velocity approaching the blade.
This behavior is followed by Silva et al. (2017), where the chord is
corrected because the term {W/[(1 − fS )VCAV ]}2 in Eq. (12) becomes
greater than 1. Fig. 7b shows the twist angle varying radially. It is
important to note that the diffuser affect the twist angle distribu-
tion, but it is not influenced by the cavitation. The power coefficient
of the present work does not have much changing compared to Vaz
and Wood (2016) (Fig. 8a). However, clearly it does to Glauert and
Durand (1963) and Silva et al. (2017). The increased power output
is about 42% when compared to the bare turbine, represented by
Glauert and Durand (1963) in Fig. 8b, for a water velocity of 2.5 m/s.
This result shows that the present optimization corrects the chord
without relevant changing in the turbine efficiency, keeping it above
Betz-Joukowsky limit if the power coefficient is based on the rotor
diameter. However, if the power coefficient is based on the exit area

Fig. 6. Effect of the speed-up ratio on the relative velocity. (a) uncorrected and (b) corrected model.

Fig. 7. Effect of the speed-up ratio on the (a) chord and (b) twist angle distributions.
148 D. do Rio Vaz et al. / Energy for Sustainable Development 45 (2018) 142–149

Fig. 8. (a) Power coefficient as a function of the tip-speed ratio, and (b) power output as a function of the water velocity.

of the diffuser, probably the maximum efficiency is going to attains Nomenclature

below Betz-Joukowsky limit. This assumption is further described Latin symbols
by Sørensen (2016). In addition, also the results point out that a a, a axial and tangential induction factors at the rotor
hydrokinetic turbine needs to be carefully designed as the cavitation aopt optimum axial induction factor
indeed increases under diffuser (or duct) effect. B number of blades
cc , c u corrected and uncorrected chord (m)
Conclusions CD drag coefficient
CL lift coefficient
An optimization procedure for the design of diffuser-augmented Ct tangential force coefficient
hydrokinetic blades free of cavitation is presented. The formulations Cn normal force coefficient
are based on the classical BET, whose the main addition being the CP power coefficient
minimum pressure coefficient as the criterion to avoid cavitation Cp pressure coefficient
on optimum hydrokinetic blades. The present method extends Vaz Cpmin minimum pressure coefficient
and Wood (2016) optimization procedure to hydrokinetic case. The dA elementary area (m2 )
model has low computational cost and easy numerical implemen- D drag(N)
tation. Also, it is demonstrated that the cavitation prediction has fS safety factor
good physical behavior. The good performance of the proposed F Prandtl tip-loss factor
optimization is made through comparison with studies developed Fn normal force (N)
by Silva et al. (2017), in which the increased chord distribution Ft tangential force (N)
can avoid cavitation. This approach suggests that the power out- g gravitational acceleration (m/s2 )
put could be increased by 42% for a water velocity about 2.5 m/s, h distance between free surface and blade section (m)
as shown in Fig. 8b, when compared with the classical Glauert’s H submergence of the turbine (m)
optimization (Glauert & Durand, 1963). In addition, when the term L lift(N)
{W/[(1 − fS )VCAV ]}2 in Eq. (12) becomes greater than 1, the chord p local pressure (Pa)
length can increase 64% compared to Vaz and Wood (2016) model. patm atmospheric pressure (Pa)
It is necessary to consider some limitations of the present model, pv vapor pressure (Pa)
such as c must depend on the diffuser efficiency, as recently r radial position at the rotor plane (m)
described by Silva et al. (2018), and this relationship requires fur- R radius of the rotor (m)
ther investigation. A new formulation for the tip loss factor must be V0 free-stream velocity (m/s)
assessed, as raised by Vaz and Wood (2016). This is because the dif- VCAV cavitation velocity (m/s)
fuser is assumed to induce a circumferentially uniform axial flow, but w total induced velocity (m/s)
no azimuthal flow through the rotor. This axial flow should reduce W relative velocity (m/s)
the tip loss due to the proximity of the blade tip to the diffuser x local-speed ratio
wall. There should be different tip loss factors for the axial and cir-
cumferential motion. Another limitation is the lack of incorporation Greek symbols
of the pressure coefficient at the diffuser outlet in the modeling. a angle of attack (rad)
This incorporation is very important for verifying the behavior of a opt optimum angle of attack (rad)
the optimization approach during cavitation, as this pressure coeffi- c diffuser speed-up ratio
cient is dependent on the velocity increase. Despite such limitations, b twist angle (rad)
the results obtained here present good behavior, demonstrating that bopt optimum twist angle (rad)
diffuser technologies can intensify the cavitation on hydrokinetic k tip-speed ratio
blades, justifying the use of new methodologies to improve current q density of the fluid (kg/m3 )
optimization procedures applied to hydro rotor design. Validation s number of cavitation
of the results with experiments or 3D numerical modeling are pre- ss solidity of the turbine
tended in future works. 0 flow angle (rad)
 D. do Rio Vaz et al. / Energy for Sustainable Development 45 (2018) 142–149 149

0opt optimum flow angle (rad) Goundar, J. N., Ahmed, M. R., & Lee, Y. H. (2012). Numerical and experimental studies
on hydrofoils for marine current turbines. Renewable Energy, 42, 173–179.
Y angular speed of the turbine (rad/s)
Hansen, M. (2008). Aerodynamics of wind turbines. Earthscan.
Hansen, M. O. L., Sorensen, N. N., & Flay, R. G. J. (2000). Effect of placing a diffuser
around a wind turbine. Wind Energy, 3, 207–213.
Acknowledgments
Jafari, S. A. H., & Kosasih, B. (2014). Flow analysis of shrouded small wind turbine with
a simple frustum diffuser with computational fluid dynamics simulations. Journal
The authors would like to thank the CNPq,CAPES, and PROPESP/ of Wind Engineering and Industrial Aerodynamics, 125, 102–110.
UFPA for financial support. Laws, N. D., & Epps, B. P. (2016). Hydrokinetic energy conversion: technology, research,
and outlook. Renewable and Sustainable Energy Reviews, 57, 1245–1259.
Ohya, Y., & Karasudani, T. (2010). A shrouded wind turbine generating high output
power with wind-lens technology. Energies, 3, 634–649.
References Philipis, D. G. (2003). PhD. thesis. An investigation on diffuser augmented wind tur-
bine design. Department of Mechanical Engineering, School of Engineering, The
University of Auckland..
Abe, K., & Ohya, Y. (2004). An investigation of flow fields around flanged diffusers using
Rio Vaz, D. A. T. D., Mesquita, A. L. A., Vaz, J. R. P., Blanco, C. J. C., & Pinho, J. T.
CFD. Journal of Wind Engineering Industrial Aerodynamics, 92, 315–330.
(2014). An extension of the Blade Element Momentum method applied to Diffuser
Adhikari, R. C., Vaz, J. R. P., & Wood, D. H. (2016). Cavitation inception in crossflow
Augmented Wind Turbines. Energy Conversion and Management, 87, 1116–1123.
hydro turbines. Energies (Basel), 9, 237.
Shinomiya, L. D. (2015). M.Sc. thesis. Design of horizontal axis hydrokinetic rotors
Anyi, M., & Kirke, B. (2010). Evaluation of small axial flow hydrokinetic turbines for
considering the effect of cavitation (Original in Portuguese: Projeto de Rotores
remote communities. Energy for Sustainable Development, 14, 110–116.
hidrocinéticos de Eixo Horizontal Considerando o Efeito de cavitação). Brazil:
Batten, W. M., Bahaj, A. S., Molland, A. F., & Chaplin, J. R. (2008). The prediction of
Graduate Program in Mechanical Engineering, Institute of Technology, Federal
the hydrodynamic performance of marine current turbines. Renewable Energy, 33,
University of Para..
1085–1096.
Silva, P. A. S. F., Rio Vaz, D. A. T. D., Britto, V., Oliveira, T. F., Vaz, J. R. P., & Brasil Junior, A.
Belloni, C. S. K., Willden, R. H. J., & Houlsby, G. T. (2017). An investigation of ducted and
C. P. (2018). A new approach for the design of diffuser-augmented hydro turbines
open-centre tidal turbines employing CFD-embedded BEM. Renewable Energy,
using the blade element momentum. Energy Conversion and Management, 165,
108, 622–634.
801–814.
Benini, E. (2004). Significance of blade element theory in performance prediction of
Silva, P. A. S. F., Shinomiya, L. D., Oliveira, T. F., Vaz, J. R. P., Mesquita, A. L. A., &
marine propellers. Ocean Engineering, 31, 957–974.
Brasil Junior, A. C. P. (2017). Analysis of cavitation for the optimized design of
Branlard, E. (2013). Master’s Thesis. Wind turbine tip-loss corrections: Review
hydrokinetic turbines using BEM. Applied Energy, 185, 1281–1291.
implementation and investigation of new models. Master of Science in Wind Energy
Sørensen, J. N. (2016). General momentum theory for horizontal axis wind turbines.
at the Technical University of Denmark..
Research topics in wind energy. Vol. Vol. 4. Springer International Publishing.
Chen, L., Ponta, F. L., & Lago, L. I. (2011). Perspectives on innovative concepts in
https://doi.org/10.1007/978-3-319-22114-4.
wind-power generation. Energy for Sustainable Development, 15, 398–410.
van Bussel, G. J. W. (1999, July). An assessment of the performance of diffuser
Dominguez, F., Achard, J. L., Zanette, J., & Corre, C. (2016). Fast power output prediction
augmented wind turbines (DAWT’S). 3th ASME/JSME Joint Fluids Engineering
for a single row of ducted cross-flow water turbines using a BEM-RANS approach.
Conference San Francisco, california, usa.
Renewable Energy, 89, 658–670.
Vaz, J. R. P., & Wood, D. H. (2016). Aerodynamic optimization of the blades of
Drela, M. (1989, June). XFOIL: An analysis and design systems for low Reynolds num-
diffuser-augmented wind turbines. Energy Conversion and Management, 123,
ber airfoils. Conference on low Reynolds number airfoil aerodynamics. University of
35–45.
Notre Dame.
Vaz, J. R. P., & Wood, D. H. (2018). Effect of the diffuser efficiency on wind turbine
Favacho, B. I., Vaz, J. R. P., Mesquita, A. L. A., Lopes, F., Moreira, A. L. S., Soeiro, N. S., &
performance. Renewable Energy, 126, 969–977.
Rocha, O. F. L. (2016). Contribution to the marine propeller hydrodynamic design
Wald, Q. R. (2006). The aerodynamics of propellers. Progress in Aerospace Sciences, 42,
for small boats in the Amazon region. Acta Amazonica, 46(1), 37–46. https://doi.
85–128.
org/10.1590/1809-4392201501723.
Wood, D. H. (2015). Maximum wind turbine performance at low tip speed ratio.
Fletcher, C. A. J. (1981). Computational analysis of diffuser-augmented wind turbines.
Submitted to Journal of Renewable and Sustainable Energy, 7, 053126. https://doi.
Energy Conversion Management, 21, 175–183.
org/10.1063/1.4934805.
Glauert, H., & Durand, W. (1963). Aerodynamic theory. In W. F. Durand (Ed.), Chapter
XI. Division l. Airplanes propellers, 4, 191-195 [reprinted]. NewYork: Dover.