Professional Documents
Culture Documents
www.elsevier.com/locate/jfranklin
Abstract
This study illustrates design optimization for multiple wind towers located at different villages in
Alaska. The towers are supported by two different types of foundations: large mat or deep piles
foundations. Initially, a reinforced concrete (RC) mat foundation was proposed. Where soil
conditions required it, a pile foundation solution was devised utilizing a 30 in thick RC mat
containing an embedded steel grillage of W18 beams and supported by 2024 in grouted or un-
grouted piles. For faster installation and lower construction cost, all-steel foundations were proposed
for these remote Alaska sites. The new all-steel design was found to reduce the natural frequencies of
the structural system due to softening the foundation. Thus, the towerfoundation system could
potentially become near-resonant with the operational frequencies of the wind turbine.
Consequently, the likelihood of structural damage or even the collapse is increased.
A detailed 3D nite-element model of the towerfoundationpile system with RC foundation was
created using SAP2000. Soil springs were included in the model based on soil properties obtained
from the geotechnical investigation. The natural frequency from the model was veried against the
tower manufacturer analytical and experimental values. When piles were used, numerous iterations
were carried out to eliminate the need for the RC and optimize the design. An optimized design was
achieved with enough separation between the natural and operational frequencies. The design
$
This paper appeared in a preliminary form in the Proceedings of The Third International Conference on
Modeling, Simulation and Applied Optimization (ICMSAO09), Sharjah, UAE, 2009.
Corresponding author. Tel.: 971 6 515 2647; fax: 971 6 515 2979.
E-mail addresses: malhamaydeh@aus.edu (M. AlHamaydeh), saifmh@gmail.com (S. Hussain).
0016-0032/$32.00 & 2010 The Franklin Institute. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.jfranklin.2010.04.013
M. AlHamaydeh, S. Hussain / Journal of the Franklin Institute 348 (2011) 14701487 1471
successfully avoids damage to the structural system, while eliminating the need for any RC in most
cases.
& 2010 The Franklin Institute. Published by Elsevier Ltd. All rights reserved.
1. Introduction
frequencies and corresponding mode shapes were investigated and compared for different
conditions accounting and neglecting the door opening and its orientation. It was
concluded that simplied FEA models can predict overall behavior with reasonable
accuracy.
2. Windstructure interaction
Slender structures obstructing steady air ow will experience aerodynamic forces in both
directions along-wind and across-wind (or drag and lift forces, respectively). If the
structure has a symmetric cross-section, the dominant force would be the drag force. When
the vortex shedding phenomenon is exhibited, the lift force increases and becomes
signicant. To illustrate this, the Bernoullis equation can be used to determine the
pressure on a structure due to steady wind as shown below
1
P rv2 1
2
where P is the pressure, n is the velocity and r is the standard air density which is
0.0761 pcf at 15 1C and 760 mm of mercury. The equivalent static pressure representing the
mean dynamic pressure is then evaluated as
Ps 0:00256V 2 2
where Ps is the static pressure in psf and V is the velocity expressed in mph as traditionally
done in the structural engineering industry [8,9].
The Drag Force (FD) and the Lift Force (FL) can be found by
FD 0:00256V 2 CD B 3
FL 0:00256V 2 CL B 4
where FD and FD are in lb/ft, B is the effective width in ft, and CD and CL are the Drag
and Lift coefcients, respectively, which depend on the obstruction shape as well as
Reynolds number, Re, which in turn is dened as the ratio of inertial forces to viscous
forces.
vB
Re 5
n
where n is the velocity in ft/s, B is the section width and n (Greek letter nu) is the
kinematic viscosity of air, which are 1.46 105 m2/s at 15 1C and 760 mm of mercury.
For a circular section, B in Eq. (5) would be taken as the diameter D. Thus, Eq. (5)
M. AlHamaydeh, S. Hussain / Journal of the Franklin Institute 348 (2011) 14701487 1473
can be rewritten as
Re 9350VD 6
where V is the velocity in mph and D is the diameter in ft.
ut
us u
F(t)
m m m
t
s
ks
where on is the undamped natural circular frequency (rad/s) can be found from
r
ke
on 11
m
The damped and undamped frequencies can be taken to be equal for low levels of
damping (less than 20% of the critical damping). Since the tower and the support are
connected in series, the equivalent stiffness ke can be calculated from the following
equation:
kks
ke 13
k ks
Eq. (14) describes the effect of the support stiffness on the overall system stiffness. When
the foundation is innitely rigid, the equivalent stiffness of the system reduces down to the
tower stiffness. Conversely, if the foundation is extremely soft, the system faces a stability
problem. For most engineered systems, the latter is highly unlikely.
The following sections will shed light on the effect of foundation modeling on the
natural frequency of the entire system.
The utilized towers are out-of-commission prefabricated models that were donated to
and/or purchased by the state of Alaska to generate electricity in the following remote
villages: Hooper Bay, Chevak, Gambell, Savoonga, Mekoryuk, Kasigluk Akula Bay,
M. AlHamaydeh, S. Hussain / Journal of the Franklin Institute 348 (2011) 14701487 1475
Toksook Bay, Danwin and Nordtank. The tower models differ in height (2346 m), weight
and wall thickness. Furthermore, the three-bladed turbines vary in mass (78127909 kg),
blade diameter (1927 m) and power output (100225 kW). All towers were supplied by
the same manufacturer, Distributed Energy Systems (DES), formerly Northern Power
Systems (NPS), while the turbines were supplied by NPS and Vestas. Fig. 2 below shows
some of the installed towers.
During early design stages, it was determined that the along-wind and across-wind
resonance phenomena will occur at relatively high frequencies, which are sufciently
separated from the natural frequencies of the system. Using Eq. (8) above, one could
calculate the frequency at which vortex shedding will occur for a typical circular tower
(Sn=0.2) with a diameter of 3 ft under the maximum design wind speed of 100 mph. The
simple calculation yields a vortex shedding frequency of 9.78 Hz, which corresponds to an
along-wind impulse frequency of 19.56 Hz. This renders both phenomena critical for the
design wind loads of the super-structure, but not an operational limitation for the turbine.
On the other hand, as the wind turbine blades start to rotate from rest, their circular
speed increases and the induced vibration frequency increases. Depending on its power
output capacity, the turbine blades rotate at maximum rotational (circular) speeds that
typically range 4560 rpm corresponding to 0.751.00 Hz. These operational frequencies
are close to the range of natural frequencies of the entire soilfoundationtowerturbine
system.
If more power output is desired, higher rotational speeds have to be accommodated. A
poor design decision would involve a maximum rotational speed that is very close to the
natural frequency of the structural system resulting in a high likelihood of resonant
amplication. In such cases, the structure would have to endure violent resonance
vibrations as the operational frequency coincides with the natural frequency. This situation
would result in very high dynamic forces, which could cause immediate damage to the
structure. Even if these dynamic forces do not exceed the structures strength capacity or
elastic limit, fatigue-induced failures could also be encountered.
A sound design would avoid allowing the operational frequency to approach the vicinity
of the natural frequency by a certain safety factor. A safety factor of 15% of the natural
frequency was recommended by the turbine vendor and adopted by the authors for this
1476 M. AlHamaydeh, S. Hussain / Journal of the Franklin Institute 348 (2011) 14701487
project. This is in agreement with the 10% separation rule-of-thumb that is typically
implemented in design [2].
It is worth mentioning here that the factor of safety mentioned above is a nominal
(conservative) one and is only used for the frequency comparison. Conservatively, it does
not reect the combined soil and structure damping which ranges 525% of the critical
damping depending on the soil type, strain level and connement pressure [14].
5. Design objective
In order to develop a sound overall structural system that meets the structural
performance requirements of the wind towers, the dynamic interaction of the supporting
soil, foundation and super-structure needs to be considered. Since the tower and turbine
are prefabricated and manufactured, once selected for a certain installation location, only
the foundation can be designed and ne-tuned in accordance with the site soil conditions
and desired system frequency.
Depending on the soil conditions, the optimum foundation system needs to be selected
(spread footing, deep piles, micro-piles, etc.). Additionally, the foundation must have
adequate stiffness in order to maximize the system natural frequency within practical
limits. A suitably stiff soilfoundationstructure system will allow for higher power output
generated by the turbines.
6. Foundation design
Based on the geotechnical conditions at different sites, two types of foundations were
selected; large spread foundation and deep piles. A 50 deep, 120 120 RC spread footing
was utilized to provide the system with vertical and lateral support as well as damping and
stiffness. Where soil conditions necessitated it, a pile foundation solution was devised
utilizing a 30 in thick mat of RC foundation embedded with a steel grillage of W18 beams
founded on 20 in grouted piles (Fig. 3 below).
After some installations were made, it was determined that the mixing and casting of
concrete in-situ is the major source of cost and difculty of construction. An all-steel
foundation was proposed for faster installation and lower cost, but such a foundation
system impacted the natural frequency and softened the system. Consequently, the
foundation design was driven by the system natural frequency rather than strength or
serviceability. Multiple solutions combining different pile sizes, grouted and un-grouted
and different beam sizes were devised. The optimum design was selected for each location
Fig. 4. (a) Tapered frame element model. (b) Meshed shell element model. (c) Arbitrary mode shapes of the shell
element model.
Table 1
Modulus of horizontal subgrade reaction at Hooper Bay location [14].
Soil depth below Modulus of horizontal Soil depth below Modulus of horizontal
surface (ft) subgrade reaction (kip/in) surface (ft) subgrade reaction (kip/in)
0 0.00 13 149.76
1 11.52 14 161.28
2 23.04 15 172.80
3 34.56 16 184.32
4 46.08 17 195.84
5 57.60 18 207.36
6 69.12 19 218.88
7 80.64 20 230.40
8 92.16 21 241.92
9 103.68 22 253.44
10 115.20 23 264.96
11 126.72 24 276.48
12 138.24 25 288.00
1478 M. AlHamaydeh, S. Hussain / Journal of the Franklin Institute 348 (2011) 14701487
based on the highest practically obtainable natural frequency and cost effectiveness of the
design.
Fig. 7. Steel tower support on top of piles which are laterally constrained by soil springs.
1480 M. AlHamaydeh, S. Hussain / Journal of the Franklin Institute 348 (2011) 14701487
optimization runs utilized the tapered frame element. Fig. 4 below shows the two different
modeling techniques.
For the foundation system, the steel grillage and piles were modeled using frame
elements with the appropriate cross-sectional properties assigned. Thick plate elements
were utilized to model the RC foundation. In order to capture the soilfoundation
structure interaction, compression-only springs were devised to mimic the soil around the
piles. Dynamic soil properties could uctuate between 30% and 80% of their nominal
values depending on the strain levels [16]. Consequently, the lower-bound properties were
conservatively used in estimating the soil stiffness. Table 1 lists the modulus of subgrade
reaction at 1 ft intervals reaching to the point of xity used for modeling on the Hooper
Bay location. The turbine mass was lumped at the hub height above the top of the tower.
The piles were meshed into 1 ft segments. The large spread footing, on the other hand, was
modeled using a 3D solid element with RC properties (Figs. 5 and 6).
Fig. 8. Meshed 3D solid element with vertical and horizontal compression-only soil springs.
The solid element was meshed into sub-elements using an intelligent algorithm
consistent with the object-based FEA modeling of the SAP2000 program. Figs. 79
show the discretization of the piles and footing with the application of the soil springs to
the meshed surfaces.
Table 2
Inuence of individual foundation components on system natural frequency.
1.28
Natural Frequency (Hz)
1.26
1.24
1.22
1.20
1.18
1.16
0 10 20 30 40 50 60 70 80 90 100
Concrete Properties (%)
9
Reduction to Natural Frequency (%)
0
0 10 20 30 40 50 60 70 80 90 100
Concrete Properties (%)
8. Results
Where piles were used, numerous iterations were carried out to eliminate the need for
the RC and optimize the design once a comfort level with the modeling technique was
reached. To acquire a sense of contribution of individual foundation components to the
overall system frequency, the components stiffness was reduced to 10% of its original
values. Table 2 illustrates the inuence of reduction individual foundation components on
the system natural frequency. As suggested by intuition, it was conrmed that the main
stiffness contribution is provided by the pile system. With each reduction increment to the
concrete properties, noticeable system frequency reductions were introduced; almost 9%
reduction upon concrete elimination. Figs. 10 and 11 depict the inuence of concrete
reduction and elimination on the system natural frequency.
0.980
0.975
Natural Frequency (Hz)
0.970
0.965
0.960
0.955
0.950
0.945
0.940
3 4 5 6 7 8 9
Pile Group Radius (ft)
Table 3
Summary of design alternatives for Hooper Bay village.
Existing foundation 1.054 1.3 Current design, with 3000 of concrete and steel grillage
New foundation 1 0.946 9.0 Original model, removed concrete
New foundation 2 0.966 7.1 Original model, removed concrete, narrowed pile group radius
from 8 to 7 ft
New foundation 3 0.979 5.9 Original model, removed concrete, narrowed pile group radius
from 8 to 6 ft
New foundation 4 0.982 5.6 Original model, removed concrete, narrowed pile group radius
from 8 to 5.5 ft
New foundation 5 0.983 5.5 Original model, removed concrete, narrowed pile group radius
from 8 to 5 ft
New foundation 6 0.981 5.7 Original model, removed concrete, narrowed pile group radius
from 8 to 4.5 ft
New foundation 7 0.978 6.0 Original model, removed concrete, narrowed pile group radius
from 8 to 4 ft
New foundation 8 0.958 7.9 Original model, removed concrete, grouted piles
New foundation 9 0.951 8.6 Original model, removed concrete, welded piles to grillage beams
New foundation 10 0.964 7.3 Original model, removed concrete, added knee braces
(HSS12 6 1/2, pinned) between piles and grillage beams
New foundation 11 0.947 8.9 Original model, removed concrete, added X-braces
(HSS12 6 1/2, pinned) between piles
New foundation 12 0.948 8.8 Original model, removed concrete, added stiffeners to piles
New foundation 13 1.019 2.0 Original model, removed concrete, narrowed pile group radius
from 8 to 5 ft, grouted piles, stiffened upper 12 ft of soil
New foundation 14 1.047 0.6 Original model, removed concrete, narrowed pile group radius
from 8 to 6 ft, 2400 (t=0.7500 ) open piles, stiffened upper 12 ft of
soil, grillage beams W36 170
New foundation 15 1.050 0.9 Original model, removed concrete, narrowed pile group radius
from 8 to 5 ft, grouted piles, stiffened upper 12 ft of soil, grillage
beams W36 300
New foundation 16 1.048 0.8 Original model, removed concrete, narrowed pile group radius
from 8 to 6.5 ft, 2400 (t=0.7500 ) open piles, stiffened upper 12 ft of
soil, grillage beams W36 170
New foundation 17 1.042 0.2 Original model, removed concrete, narrowed pile group radius
from 8 to 6.5 ft, 2400 (t=0.7500 ) open piles, stiffened upper 12 ft of
soil, grillage beams W33 130
New foundation 18 1.049 0.9 Original model, removed concrete, narrowed pile group radius
from 8 to 7 ft, 2400 (t=0.75) open piles, stiffened upper 12 ft of
soil, grillage beams W36 170
New foundation 19 1.042 0.1 Original model, removed concrete, narrowed pile group radius
from 8 to 7 ft, 2400 (t=0.7500 ) open piles, stiffened upper 12 ft of
soil, grillage beams W33 130
New foundation 20 1.050 0.9 Original model, removed concrete, narrowed pile group radius
from 8 to 7.5 ft, 2400 (t=0.7500 ) open piles, stiffened upper 12 ft of
soil, grillage beams W36 170
New foundation 21 1.041 0.1 Original model, removed concrete, narrowed pile group radius
from 8 to 7.5 ft, 2400 (t=0.7500 ) open piles, stiffened upper 12 ft of
soil, grillage beams W33 130
New foundation 22 1.043 0.3 Original model, removed concrete, pile group radius 8 ft, 2400
(t=0.7500 ) open piles, stiffened upper 12 ft of soil, grillage beams
W36 135
M. AlHamaydeh, S. Hussain / Journal of the Franklin Institute 348 (2011) 14701487 1485
damage to the structural system. The optimization eliminated the need for any RC
encasement to the steel foundation or grouting to the piles in many cases. Table 3
illustrates the comparison of the different considered designs for one of the Hooper Bay
village locations.
In most cases, an optimized foundation system design for a particular site was also
found to be satisfactory for other locations. Thus, a small library of universally applicable
standard designs was compiled in an effort to keep the fabrication cost low. Table 4
Table 4
Final design summary for Savoonga and Mekoryuk villages.
Savoonga
Tower height 29 m
Turbine C.G. 1.28 m (50 in) Above tower
Turbineblades mass 7812 kg-mass (535.292 slugs)
Foundation beams W36 170
Pile section 24 in steel pipe, 3/4 in thick
Number of piles 6
Point of xity 1811 ft below soil surface
Modulus of horizontal subgrade reaction 19319 kip/in
System natural frequency 1.128 Hz
Recommended maximum rpmn 57 rpm
Mekoryuk
Tower height 29 m
Turbine C.G. 1.28 m (50 in) Above tower
Turbineblades mass 7812 kg-mass (535.292 slugs)
Foundation beams W36 170
Pile section 24 in steel pipe, 3/4 in thick
Number of piles 6
Point of xity 30 ft Below soil surface
Modulus of horizontal subgrade reaction 191070 kip/in
System natural frequency 1.148 Hz
Recommended maximum rpma 58 rpm
a
The recommended rpm incorporates a 15% safety factor.
summarizes the nal design for two of the tower locations and demonstrates how one
optimized design was found to be adequate at an another location with different
geotechnical conditions.
Fig. 14 below shows the optimized all-steel tower support foundation in its nal state
just before eld installation.
9. Conclusions
References
[1] M. Al Satari, S. Hussain, Vibration-based wind turbine tower foundation design utilizing soilfoundation
structure interaction, in: Proceedings of The Third International Conference on Modeling, Simulation and
Applied Optimization (ICMSAO09), Sharjah, UAE 2009.
[2] International Electrotechnical Commission 61400-01, Wind Turbine Generator SystemsPart 1: Design
requirements, 2005.
[3] Bonnett Danny, Wind turbine foundations loading, dynamics and design, Structural Engineer 83 (3)
(February 1, 2005) 4145.
[4] Gill, Alistair; Fraser, Ramsay, Challenges in the design of an offshore wind turbine foundation for arctic
conditions, in: Proceedings of the International Conference on Offshore Mechanics and Arctic
EngineeringOMAE, v 4, pp. 507515, 2002.
[5] Fei, Chaoyang; Wang, Nan; Zhou, Bo; Chen, Changzheng, Dynamic performance investigation for large-scale
wind turbine tower, in: Proceedings of the Eighth International Conference on Electrical Machines and
Systems (ICEMS), v 2, (2005) pp. 996999.
[6] M.B. Zaaijer, Foundation modeling to assess dynamic behavior of offshore wind turbines, Applied Ocean
Research 28 (1) (February 2006) 4557.
[7] Xu, Yan; Sun, Wenlei; Zhou, Jianping, Static and dynamic analysis of wind turbine tower structure, Advanced
Materials Research, v 33-37 PART 2, 2008 pp. 1169-1174, Advances in Fracture and Materials
BehaviorSelected, peer reviewed papers of the Seventh International Conference on Fracture and Strength
of Solids (FEOFS2007).
[8] American Society of Civil Engineers (ASCE), ASCE 7-05: Minimum Design Loads for Buildings and Other
Structures, Reston, Virginia, 2005.
[9] International Code Council (ICC), in: International Building Code, 2009 edition, Falls Church, Virginia,
2009.
[10] E. Simiu, R. Scanlan, in: Wind Effects on Structures: Fundamentals and Applications to Design, third
edition, John Wiley & Sons, New York, 1996.
[11] J. Mander, S. Chen, K. Shah, A. Madan, Investigation of light pole base integrity report, research project
funded by the Erie County Department of Public Works, Erie County, New York, 1992.
[12] B. Taranath, in: Steel, Concrete, & Composite Design of Tall Buildings, second edition, McGraw-Hill, New
York, 1997 pp 123124.
[13] A. Chopra, in: Dynamics of Structures: Theory and Applications to Earthquake Engineering, rst edition,
Prentice Hall, Upper Saddle River, New Jersey, 1995.
M. AlHamaydeh, S. Hussain / Journal of the Franklin Institute 348 (2011) 14701487 1487
[14] S. Teachavorasinskun, P. Thongchim, P. Lukkunaprasit, Shear modulus and damping of soft Bangkok
Clays, Canadian Geotechnical Journal v 39 (2002) 12011208.
[15] R. Cook, D. Malkus, M. Plesha, in: Concepts and Applications of Finite Element Analysis, John Wiley &
Sons, New York, 1989 pp 553582.
[16] Golder Associates, Inc., Geotechnical Reports, Anchorage, Alaska, 2005.