347 14-Ait PDF
347 14-Ait PDF
347 14-Ait PDF
Keywords
0 -3,89E-16
0 5 10 15 20
Cl Cd
Fig.4. Lift and Drag coefficients versus the angle of attack for
the NACA 63-415 section as computed with XFoil
The turbine rotor described and tested by Bahaj et al. [6] (7)
is considered. For this rotor a considerable set of
experimental data obtained from cavitation tunnel and Where is the duct drag coefficient and is the rotor
towing tank tests is available in the literature [6]. The thrust coefficient. In the case of un-ducted turbine, the
turbine is a three-bladed turbine with NACA 63-415 duct drag coefficient is equal to zero.
sections. The standard geometry has a pitch angle at the
blade root equal to 15°, corresponding to a 0° set angle,
the blade set angle is the angle at the tip of the blade 0,8
(pitch). In the present work, 0°, 5°, 10° and 13° blade set
angle were considered. The rotor hub can be easily
modelled but in most cases it does not influence the 0,6
hydrodynamic coefficients so in the results presented here,
we do not simulate its presence. 0,4
A view of the blade mesh and the turbine rotor geometry Panel Method
Cp Cav. Tunnel 1.40 m/s
is shown in figure 3. analytical
0,2
0
2 4 TSR 6 8 10
10
0
0,2 0,4 0,6 0,8 1
-10
Chord r/R
4. Ducted Turbine
Augmentation channels induce a sub-reference pressure
within a constrained area and thereby increase the flow
velocity. If a turbine is placed in such a channel, the flow
velocity around the rotor is higher than the current
velocity. Since the potential power is proportional to the
cube of the inlet velocity, the expected gain can be very
important.
These structures are fixed at the periphery of the rotor to
increase the power extracted by the marine current
turbine, geometry of marine current turbines facilitates the Fig.11. Distribution of the pressure coefficient on the duct at
introduction of this type of device. These systems create a TSR=7.5
funnelling effect which increases the flow through the
rotor. Drag exerted by the fluid on the duct is translating Figure 12 shows the evolution of the power coefficient Cp
on a depression at the output of the duct which is at the of the ducted turbine with a NACA4424 profile and the
origin of the suction phenomenon [13]. bare turbine with 0.950m rotor diameter. The numerical
Figure 10 shows the evolution of the power coefficient as results show, for the same overall area, that the ducted
a function of the rotor drag coefficient and the duct drag turbine produces more power than the bare turbine. The
coefficient. The analytical expression of the power TSR are not the same since the action of the duct shifts the
coefficient in the presence of a duct is the equation (7). advance parameter ( at versus
at ). If these findings are confirmed, it
means that it is advisable to install water turbines
equipped with a duct.
0,8
0,6
Cp
0,4
0,2
0
2 4 TSR 6 8 10
-0,2
Panel Method with duct
Fig.10. Evolution of the power coefficient as function of the Analytical Cp with duct
rotor drag coefficient and the duct drag coefficient Panel Method without duct
Analytical Cp without duct
In addition, the aim of this work is also to analyze the Fig.12. Comparison of power coefficient Cp predicted by Panel
Method between ducted and un-ducted turbine for the same
effect of a 0.950 m diameter and 0.4m chord duct with overall
Conclusion and future work [4] H. W. M. Hoeijmakers, “Panel Methods for Aerodynamic
Analysis and Design”, AGARD Report 783. 1992, pp. 5.1-
5.47.
An academic panel code based on potential flow theory [5] J. M. Laurens, S. Moyne and F. Deniset, “A BEM Method
has been used to assess the hydrodynamic performance of for the Hydrodynamic Analysis of Fishing Boats Propulsive
a marine current turbine with and without a duct. The Systems”, in Proc. ISFVEE 2012, pp.1-7.
numerical results for the bare turbine have been compared [6] A.S. Bahaj, W. M. J. Batten, G. McCann, “Experimental
with the experimental results reported in reference [6]. Verifications of Numerical Predictions for the
Although the trend is the same, the numerical results Hydrodynamic Performance of Horizontal Axis Marine
present significantly lower values for the power Current Turbines”, Renewable Energy. 2007, Vol. 32, pp.
2479-2490
coefficient. Baltazar and Falcao de Campos [8] obtained [7] W. M. J. Batten, A. S. Bahaj, A. F. Molland, J. R. Chaplin,
the same results with a very similar method and they “Experimentally Validated Numerical Method for The
propose to correct the section drag to account for flow Hydrodynamic Design of Horizontal Axis Tidal Turbines”,
separation. From the spanwise distribution of sections Ocean Engineering. 2006, Vol. 34, pp. 1013-1020.
angles, it appears that flow separation must occur even for [8] J. Baltazar, J. A. C. Falcão de Campos, “Unsteady Analysis
configurations with high power coefficient values. Flow of a Horizontal Axis Marine Current Turbine in Yawed
separation and cavitation being closely linked, serious Inflow Conditions with a Panel Method”, in Proc. SMP2009,
investigation has to be carried out to decide whether it is pp.1-9.
advisable to design a water turbine presenting flow [9] I. Afgan, J. McNaughton, S. Rolfo, D. D. Apsley, T. Stallard,
and P. Stansby, “Turbulent Flow and Loading on a Tidal
separation even in normal operational mode. Simulations Stream Turbine by LES and RANS”, International Journal of
with a Navier-Stokes solver are needed as long as the Heat and Fluid Flow. 2013, Vol 43, pp.96-108.
turbulence model is capable of predicting the correct [10] J. McNaughon, “ Turbulence Modelling in the Near-Field of
hydrodynamic forces in presence of flow separation which an Axial flow Tidal Turbine in Code-Saturne”, First Year
is known to be a challenging problem. Experimental Report, 2010, pp.1-15.
results with flow separation monitoring are also needed. [11] P. Liu, “A Computational Hydrodynamics Method for
Horizontal Axis Turbine, Panel Method Modeling Migration
In the course of our study, we questioned the upper limit from Propulsion to Turbine Energy”, Energy. 2010, Vol. 35,
given by the GGS model [2] which is half the Betz limit. pp. 2843-2851.
We found that the GGS model is based on a wrong [12] I. H. Abbott and A. E. Von Doenhoff, Theory of Wing
assumption and serious contradiction are embedded in the Sections, Dover Edition, New York (1959), pp. 110-187 and
proof. Surprisingly the paper has been cited more than a 536-537.
hundred times and only one author suspected something [13] B. Multon, Energies Marines Renouvelables, Lavoisier
Edition, Paris (2011), pp. 277-281.
was wrong and no one investigated any further.
Finally, the results of our simulations with the presence of
a duct show that the addition of this appendix produces a
very significant increase in power outlet with a same
overall area.
The arguments given against the addition of a duct are
usually coming from the structural point of view. Our
future line of investigation will therefore concern the
material and structure behaviour of the ducted water
turbine.
Acknowledgement
References