![Reduced stresses [MPa]](https://www.researchgate.net/profile/Pawel-Dymarski/publication/276512078/figure/fig3/AS:735597392060419@1552391493648/Reduced-stresses-MPa_Q320.jpg)
![Reduced stresses [MPa]](https://www.researchgate.net/profile/Pawel-Dymarski/publication/276512078/figure/fig4/AS:735597392035849@1552391493715/Reduced-stresses-MPa_Q320.jpg)
Content uploaded by Paweł Dymarski
Author content
All content in this area was uploaded by Paweł Dymarski on Mar 12, 2019
Content may be subject to copyright.
Available via license: CC BY-NC-ND 3.0
Content may be subject to copyright.
POLISH MARITIME RESEARCH, No 1/201536
POLISH MARITIME RESEARCH 1(85) 2015 Vol. 22; pp. 36-46
10.1515/pom r-2015- 0006
DESIGN AND STRENGTH CALCULATIONS OF THE TRIPOD
SUPPORT STRUCTURE FOR OFFSHORE POWER PLANT
C. Dymarski, Prof.
P. Dymarski, Ph. D.
J. Żywick i, Ms. C.
Gdańsk University of Technology, Poland
AB ST R AC T
e support structure being the object of the analysis presented in the article is Tripod. According to the adopted
assumptions, it is a foundation gravitationally set in the water region of 60 m in depth, not xed to the seabed, which
can be used for installing a 7MW wind turbine. Due to the lack of substantial information on designing and strength
calculations of such types of structures in the world literature, authors have made an attempt to solve this problem
within the framework of the abovementioned project. In the performed calculations all basic loads acting on the
structure were taken into account, including: the self mass of the structure, the masses of the ballast, the tower and
the turbine, as well as hydrostatic forces, and aero- and hydrodynamic forces acting on the entire object in extreme
operating conditions.
Keywords: oshore wind turbine, support structure, Morison equation, FEM
INTRODUCTION
e here presented work has been done withi n the framework
of the research project AQUILO entitled “Development of
methods for the selection of the type of support structure for
oshore wind turbine in Polish sea areas”.
The purpose of the task in which this research was
implemented was to design a support structure in a given
area of the Polish economic zone on the Baltic Sea. Within
this project four types of structures were analysed: a gravity
base, a tripod with pile foundation, a gravity tripod, and
a deep water monopile.
Fig. 1a. ree types of support structures analyzed within the framework of the
project - Gravity base
Fig. 1b. ree types of support structures analyzed within the framework of the
project - Gravity tripod
e support structure being the object of the analysis
presented in the article is Tripod. According to the adopted
assumptions, it is a gravitationally set foundation, not xed
to the seabed, which can be used for installing a 7MW wind
turbine in the water region of 60 m in depth. Due to the lack
Unauthenticated
Download Date | 3/12/19 12:47 PM
POLISH MARITIME RESEARCH, No 1/2015
37
of substantial information on designing and calculations of
such types of structures in the world literature, authors have
made an attempt to solve this problem within the framework
of the abovementioned project. e object was assumed to be
loaded with: the self weight, wind pressure, outer hydrostatic
pressure (seawater), inner hydrostatic pressure (liquid ballast),
and the hydrodynamic action of the environment.
Fig. 1c. ree types of support structures analyzed within the framework of the
project - Monopile
CALCULATION OF HYDRODYNAMIC
FORCES
M
e total hydromechanical force which acts on a motionless
body placed in the unsteady, unidirectional ow of uid can
by expressed by [1]:
(1)
where A
p
is the projected area of the body on the plane
normal to the direction of the ow V
b
, is the volume of the
body, CD and C’aare the drag and added mass coecients,
U is the velocity, and ρ is the water density.
Coecients C
D
and C’
a
depend on time, geometry of the
body, Reynolds number Rn, and parameters describing the
history of the ow (for example the amplitude and time interval
of velocity variation). In practice, they are plotted as functions
of the Keulegan-Carpenter number K
C
and the so called β=
=Rn / KC coecient proposed by Sarpkaya [1].
e Morison equation is a simplied version of equation (1),
obtained aer assuming that the term dU/dt in the equation
can be approximated by U/t , hence:
(2)
where:
CM = 1 + Ca, and Ca is the time averaged value of C’a.
e above equation is used for calculations of cylindrical
shapes. When the motion of uid particles caused by waves is
to be analysed, this equation can be applied when the cylinder
diameter is not greater than about 20% of the wave length λ.
Coecients CM and CD can be obtained from model tests.
Sarpkaya [1,2] carried out a systematic study, based on which
the characteristics of coecients CM and CD were derived for
a cylindrical shape as functions of KC, β, and relative roughness
k
r
/ D. Values of these coecients for other geometries are also
available in the literature [1,3], Fig 2, 3.
Fig. 2. Comparison of inertia coecient CM for six rough cylinders,
kr/D=50. Source: Sarpkaya [2].
Fig. 3. Comparison of drag coecient CD for six rough cylinders,
kr/D=50. Source: Sarpkaya [2].
G M
-
.
In the previously presented equations an assumption is
made that the velocity vector is perpendicular to t he axis of the
Unauthenticated
Download Date | 3/12/19 12:47 PM
POLISH MARITIME RESEARCH, No 1/201538
cylinder and the velocity eld is uniform. In the general case,
for example, when dealing with the velocity eld appearing
in the wave motion, the velocity eld is not uniform, and
the direction of ow (and acceleration) is subject to changes.
e geometry of the structure varies as well. erefore the
force acting on the section of a structure element needs to be
determined using the general formula:
(3)
e velocity vector normal to the longitudinal axis of the
element of structure is given by:
Un = U - (U es)es , where: U is the velocity vector, es is the
unit vector tangent to the axis of the structure, ∆s is the length
of the section of the element in which the hydrodynamic force
is calculated, D is the diameter of the section (or the longest
diagonal if the section is a polygon). It is assumed that CM and
CD coecients are functions of KC, β, and of the geometr y of
the analysed section (and adjacent sections).
e value of the force acting on the element can be therefore
determined as the integral:
(4)
D
So far, we have assumed that the velocity eld is a known
quantity. However, determining the velocity eld is not a simple
task. e Maritime Institute in Gdansk analysed the hydro-
meteorological data for a particular sea area and performed
statistical calculations, based on which a set of basic parameters
of the waves and sea current was obtained [5].
In the project, the assumed lifetime of the structure was 30
years, with an optional extension for further 20 years, therefore,
the analysis took into account parameters of a violent storm,
the one which happens once in 50 years. Velocities of the
sea currents in the Baltic are relatively low. To calculate the
desired parameters, it was assumed in the project that the
speed of the sea current at the surface is U
curr
(0) = 0.45 m/s,
which is the velocity that appears once in 50 years in the water
region of interest.
Figure 4 presents the wave spectrum distribution for
a specic set of data, whereas Figure 5 presents the function
of the sea current velocity.
e resulting velocity eld is a sum of the sea current velocity
and the velocity eld due to waves:
(5)
Fig. 4. Wave spectrum JONSWAP for the 50-year storm: Tp=11.3s;
HS=9.01 m; γ=4.14
Fig. 5. Approximation of the current velocity prole for the 50-year storm [5]
C
e wave spectrum with the parameters described in the
above Section corresponds to the storm duration of 3 hours.
For this spectrum three wave functions were randomly chosen,
each of one hour duration. Additionally, the calculations took
into account the presence of the sea current.
e results of calculations for the wave at which the
maximum load of the structure was observed are shown below.
Figure 6 shows the bending moment relative to the bottom of
the structure (z = -60 m) as a function of time. It was assumed
that the maximum load is such that the bending moment of the
structure reaches its maximum. For the here presented results
the maximum stress occurs at time t = 3450.5 s.
Fig. 6. e total bending moment My (at z=-60m) induced on the structure by
the wave and current
Unauthenticated
Download Date | 3/12/19 12:47 PM
POLISH MARITIME RESEARCH, No 1/2015
39
DESCRIPTION OF THE INITIAL
STRUCTURE
e foundation was designed as a thin-wall structure, the
shape of which resembles a tripod. Similar marine structures
are already in operation, but they dier by foundation methods.
For the time being, they have been used in more shallow waters
and their legs were xed to the seabed using piles. e here
presented structure is intended to be used in conditions
characteristic for the Baltic Sea, in its southern region at an
approximate depth of 60 m, and is assumed to be founded on
the seabed only with the aid of gravitational forces. at is
why the shape of the lower part of the structure is designed
in such a way that, aer proper ballasting of the inside, the
foundation is ready to carry maximum loads taking place
during its operation, with no shi with respect to the seabed.
e support structure comprises a column divided into two
basic parts. e upper part has the shape of a cone, with
a ring on one end for mounting the wind turbine tower, while
the lower part has the shape of a cylinder of 8 m in diameter.
ree legs, evenly distributed by 120 degrees and inclined at
60 degrees to the vertical, are welded directly to the cylinder.
e lower part of each of these legs is shaped as an elliptical
cylinder with vertical walls, frequently referred to as the “hoof”
or foot. e hoofs were connected together using horizontal
pipes of about 1m in diameter. e footbase area is a circle of
about 30 m in radius. e overall dimensions and thickness
of the sheeting for preliminary calculations are given in Fig. 7.
Fig. 7. Basic geometric parameters of the support structure. [6]
(a) (b)
LOADS AND OPERATING CONDITIONS OF
THE SUPPORT STRUCTURE
e below described loads and conditions of operation of
the support structure are the same for each stage of structure
geometry modications described in the article. e presented
results of consecutive simulations were obtained for these
assumptions. As a lready mentioned, the structure was assumed
to be founded in the Baltic Sea area at the depth of 60 m. e
upper ange to which the column will be xed is situated 15
m over the water surface. e loads of the structure coming
from environmental conditions were generated during
a hydromechanical simulation taking into account the
conditions corresponding to the most violent storm in recent
50 years on the Baltic Sea. In these calculations, the worst
conditions for the structure were assumed which take place
when the wind loads, i.e. the thrust and the torque, mostly
generated on the turbine, and the hydromechanical forces act
along the same Y-direction.
Technical parameters of the turbine which were used for
calculating foundation loads are collected in Tablel 1.
Based on the hydromechanical simulation, the distribution
of horizontal loads coming from the sea current, waves, and the
wind was determined as the function of the height of the object.
For calculating purposes, the model was divided into 2-metre
long segments, and the averaged horizontal load acting on each
segment was calculated. e zero level was assumed at the
bottom of the structure, while the water surface corresponded
to the height of 60 m. e distribution of continuous load
along the height of the structure is shown in Fig. 8. e forces
coming from the aerodynamic drag of the tower, the thrust
generated by the turbine, and the self weight were reduced to
Tab. 1: Technical parameters of the turbine
Unauthenticated
Download Date | 3/12/19 12:47 PM
POLISH MARITIME RESEARCH, No 1/201540
the upper ange of the support structure. e following load
values were obtained: torque – 260 MNm, lateral force – 3,23
MN, vertical force – 9,2 MN.
e hydrostatic pressures of the sea water and the pulp
used for ballasting the foundation act in the structure in the
following way:
• Sea water acts on the structure along the distance
from the wave bottom (h = -10 m, below the average sea
water level) to the seabed (h = -60 m). The density of sea
water was assumed as equal to 1026 kg/m3 (marked blue
in Fig. 9)
• The ballast (pulp) acts on the inner surface of the
structure up to the average water surface level. The density
of the pulp was assumed as equal to 1700 kg/m3. (marked
red in Fig. 9)
e assumed model of load ai ms at simulating the conditions
when, in heavy seaway, the wave bottom reaches the depth of
10 m, and the hydrostatic pressure of the ballast inside the
structure is not balanced by the pressure coming from the
sea water. e distribution of pressures along the structure
is shown in Fig. 9. Fig. 8. Continuous load acting at given height.
Fig. 9. Hydrostatic load acting on the structure
DISCRETE MODEL
e lump model of the support structure shown in Fig.
7 was created in the Autodesk Inventor package. en, the
programme HyperMesh was used to obtain midsurfaces, which
made the basis for generating a grid with the aid of P-Shell
elements: second-order tetragons and triangles taken from the
library of the Radios (Optistruct) solver. e elements used in
the rst iteration calculations are collected in Table 2.
e connection areas between the support structure and
the turbine tower were modelled using perfect ly rigid elements
connected together in the central node, to which loads were
applied in the form of the torque and forces coming from the
action of the wind on the higher situated tower and turbine,
and from their self weight - Fig. 10.
Tab. 2: List of elements
Unauthenticated
Download Date | 3/12/19 12:47 PM
POLISH MARITIME RESEARCH, No 1/2015
41
Fig. 10. Central node „ange”
MODEL OF THE MATERIAL
e assumed linear-elastic model of the material revealed
the following properties:
• Young’s modulus E = 210 GPa
• Poisson ratio ν = 0.3
• density ρ = 7860 kg/m3
LOAD APPLICATION
According to the above description, the FEM model was
loaded at the central point with a negative torque of about
260 MNm along the X-direction and the lateral force of 3.23
MN acting along the Y-direction.
e loads from the masses of the wind tower column (458
t), and the nacelle and rotor (480 t) were applied as the 9.2 MN
force to the node on the upper ange.
e gravitational acceleration a = 9.81m/s2 with the
negative sense was applied along the Z-direction. According
to the assumptions described in the previous Section, the
hydrodynamic load of the structure was applied to nodes
along the Y-direction. e grid of the model was divided into
horizontal segments of 2 m in height, and then the numbers
of nodes composing particular segments were counted. Each
of these nodes was loaded with a force which was calculated
by dividing the total load of the given segment obtained
from CFD simulation by the number of nodes. A decision
to make use of the above model was dictated by the fact that
hydrodynamic pressures are much smaller than hydrostatic
pressures acting on the foundation. eir action generates
the bending moment, which is a remarkable load for the
structure. e adopted load distribution does not aect much
the local strength of the structure, and remarkably facilitates
data preparation for calculations. e hydrostatic pressure
was applied along the normal direction to the seabed and the
sheeting, according to the assumptions shown in Fig. 9.
FEM CALCULATIONS
e structure strength calculations were performed using
the Finite Element Method (FEM) and the HyperWorks v12
package produced by Alta ir. e programme Hy perMesh, which
in authors’ opinion is an excellent tool for grid preparation
and imposing boundary conditions to such a large surface
model, was used as preprocessor. Linear calculations were
performed using the solver Optistruct, while the results were
displayed using the postprocessor HyperView. e simulation
was performed on PC equipped with a 4-core, 64-bit processor
Intel I7 2.30 GHz, 16 GB of RAM DDR3, and the hard disc SSD
840 PRO. e calculations for the structure at nal stages of
geometry modications, in which the number of nite surface
elements was approximately equal to 1 million, took about
2.5 hours.
STRENGTH CALCULATIONS OF THE
INITIAL STRUCTURE
For the purpose of preliminary analyses, a structure was
worked out without inner stiening elements in order to detect
places which would require installation of additional structural
elements. A motivation for this decision was to avoid excessive
dimensioning of the structure. In the rst simulation the model
was xed at the seabed by removing the ability to move in
X,Y,Z-directions from the nodes composing the bottom of
the structure.
Figures 11 and 12 show contour maps presenting the
distributions of stresses [MPa] and deformations [mm] in
the initial structure.
Fig. 11 Reduced stresses [MPa]
Fig. 12 Elastic deformations [mm]
Unauthenticated
Download Date | 3/12/19 12:47 PM
POLISH MARITIME RESEARCH, No 1/201542
As had been expected, the largest stresses, amounting to
near 1600 MPa, were observed in the connection areas of
the legs with the outer sheeting of the cylindrical part of the
column. e maximal load generated displacements along the
Y-direction were recorded on the upper ange and amounted
to about 400 mm. e displacement along the Z-direction, of
about 600 mm, was caused by the action of the ballast pressure
on the unstiened plate of the cylinder bottom. e mass of
the structure was 1307 t.
Due to very high stresses recorded in the analyses structure,
its geometry had to be modied in the regions where maximal
stresses were observed.
MODIFICATIONS
• e obtained results have led to the conclusion that the
areas in which the legs are connected with the column should
be remarkably reinforced. A decision was made to shi the
upper ends of the legs to the inside of the cylinder and link
them together on common wedges. During further structure
geometry changes, rings and special T-type (600x30x200x30)
stieners were introduced to the inner surfaces of the legs in
the area where they cross the cylindrical part of the column.
Stieners situated closer to the axis of the structure were
linked together, while the height of the remaining stieners
was reduced, as can be seen in the enlarged section of Fig. 13.
• e plate of 50 mm in thickness which constituted
the cylinder bottom was radially stiened using trusses of
dimensions T1200x30x330x30.
• In order to perform the simulation for most
unfavourable conditions assuming the absence of friction
forces between the legs, moving along with the accompanying
ground, and the seabed, the method of model supporting
was changed. e ability to move along X,Y,Z-directions was
removed from the nodes composing the base of this leg which
was most pressed to the seabed, while the remaining two legs
were allowed to move in the X-Y plane. is simplication went
in a safe direction, as in fact, part of normal forces carried
by horizontal pipes of the structure should be taken over by
friction forces acting between the legs and the bottom of the
water region.
e results of the simulation performed for the support
structure modied in the above way are shown in Fig. 13. e
scale was selected in such a way that the red colour indicates
stresses exceeding 300 MPa.
e introduced geometry changes resulted in the reduction
of maximal stresses to about 1100 MPa, also the area of
occurrence of critical stresses (σr > 300 MPa) became smaller.
e displacement of the top of the support structure column
also decreased and amounted to 277 mm. e change of
boundary conditions at the contact with the basis to enable
leg spreading resulted in stretching of the horizontal pipes,
which provoked stresses amounting to about 900 MPa in
the area of their contact with the hoops. e reason for such
high stresses was the absence of stieners in the pipe/hoop
connection region.
Fig. 13. Reduced stresses [MPa]
e displacement of the legs in the X-Y plane was equal to
about 25 mm. e mass of the analysed structure was 1416 t.
An attempt to decrease the stresses in most vulnerable places
of the structure has led to its further modication.
• Another ring was installed in the leg/column
connection region, on the inner surface of each leg, and the
number of inner stieners along their perimeter was increased.
• e structure which closes the cylinder bottom
was reinforced with extra bulb ats 430x20, arranged into
a shape of concentric hexagons, which, along with radial trusses
composed a grate.
• Triangular vertical plates of 30 mm in thickness were
added to reinforce the upper sheeting of the cylindrical part.
ese plates played a role of gussets improving the stress
distribution in the leg/cylinder connection area.
e adopted geometry changes remarkably decreased the
stresses and the area of their occurrence, without changing
the thick ness of the material used for outer sheeting, as shown
in Fig. 14.
Fig. 14. Reduced stresses [MPa]
Unauthenticated
Download Date | 3/12/19 12:47 PM
POLISH MARITIME RESEARCH, No 1/2015
43
e structure of the lower parts of the legs (hoops) in which
high stresses had been observed as a result of the action of
forces taken over from the stretched horizontal pipes was
modied in the following way :
• Lower parts of the legs were remarkably modied.
e sheeting of the bottom which had been earlier unstiened
was reinforced with trusses (T1000x30x300x30) and bulb ats
430x20.
• Two ring stieners (T500x30x200x30) were added
in the horizontal plane.
• Ends of the horizontal pipes were moved to the insides
of the hoops by about 1 m and additionally linked with the
stieners.
• e diameter of the pipes was increased from z 1m
to 2 m.
e stress distribution resulting from the above geometry
changes introduced to the lower part of the structure is shown
in Fig. 15.
Fig. 15. Reduced stresses [MPa]
MODEL TAKING INTO ACCOUNT SEABED
PLASTICITY
e next stage of modelling oriented on improving the
accuracy of the simulation took into account the plasticity
of the seabed. Solving this problem required the knowledge
about physical properties of the seabed structure in the water
region of interest. It is noteworthy that this information is,
generally, available, as one of the initial stages of investments
projects of this type is examination of the seabed structure
in the area of future foundation of the supporting structures.
However at present, due to the lack of access to the results
of this examination, the simulation was performed for the
seabed stiness taken from the standard PN-80/B-03040
“Foundations and supporting structures for machines” [7].
e ground stiness coecient Cz = 40 MPa/m was chosen
from Table 1 in the above standard, which corresponds to
the “low-stiness II category ground, silty sands, hydrated”.
e plasticity of the ground was modelled using spring type
elements. e nodes situated at the bottom of legs were copied
and moved apart by the distance of 1 m along the Z-direction,
and the elements of spring type “Celas_1” from the library of
the solver Radios, which reveal stiness at the longitudinal
direction, were inserted between them. e stiness of the
individual spring was calculated using the following formula:
(6)
where:
Ki – stiness of a single spring [N/m];
Cz = 40 [MPa/m] – ground stiness coecient, according
to [7];
A = 308,9 [m2] – total area of legs of the structure;
i = 134138 [-] – number of springs, equal to the number
of nodes on structure’s legs.
en, certain degrees of freedom were removed from the
lower nodes of the springs in such a way that leg spreading
could still be taken into account. In case of the leg on which the
load Y acts (the leg does not move horizontally with respect
to the seabed), the upper nodes of the springs were deprived
of ability to move in the XY plane, while they were still able to
move in the Z-direction. e motion of the nodes situated at the
lower ends of the springs supporting the structure was blocked
in all three directions. is solution provided opportunities for
analysing the eect of the interaction between the seabed and
the base of the structure. Some changes were also introduced
to the geometry of the structure. e sheeting of the side walls
of the hoops in the area of connections with the horizontal
pipes was reinforced with four vertical stieners of T type
(300x30x200x30), while the pipe segments situated inside the
hooks were reinforced with gussets of 30mm in thickness.
e cylinder sheeting between the legs coming into it was
reinforced with additional horizontal ring segments. e way
of model xing and the results of simulation calculations for
the structure with the introduced changes are shown in Figs.
16 and 17.
As had been expected, the obtained results have revealed the
eect of seabed plasticity on the deformations and stresses of
the structure, in particular in the contacting areas of the main
structure elements in which the maximal stress increment
amounted to about 6 %. Additionally, the forces exerted by
the legs of the structure on the bottom of the water region were
calculated. For the foot on the direction of which the load acts
the force was equal to 1,02 MPa, while for the remaining two
legs it was 0,68 MPa for each. e analysis of the obtained
results proves their qualitative correctness, which provides
good opportunities for further perfecting of the examined
model.
Unauthenticated
Download Date | 3/12/19 12:47 PM
POLISH MARITIME RESEARCH, No 1/201544
Fig. 16. Model xing, reduced stresses [MPa]
Fig. 17. „Cylinder”, reduced stresses [MPa]
FINAL STAGE OF MODIFICATION.
CHECKING CALCULATIONS
e last modication has led to the structure which
meets the assumed strength criteria. However, due to high
stresses still existing in the leg/cylinder connection area, the
thickness of the sheeting of the legs above the last upper ring
was increased to 40 mm. Also a decision was made to make
the central part of the leg sheeting, in which the stress level
was relatively low, using a thinner, 24 mm sheet, additionally
reinforced with three rings T300x24x200x20 to prevent
possible stability loss (buckling) of the sheeting. Positions of
these rings are marked in Fig. 18. Moreover, some changes were
introduced in the hoop area to reduce stresses. Aer analysing
the results taking into account the seabed elasticity, the bottom
plate of the structure was reinforced with additional radially
distributed bulb ats 370x16. e sheets of the side sheeting
in the horizontal pipe entry areas, between the outer vertical
stieners, were thickened to 40 mm. e ending parts of the
horizontal pipes which entered into the hoops were made of
40-mm thick sheet. Additionally, to stien their free edges,
inner rings of 30 mm in thick ness were mounted on the ends of
these pipes. e thickness of the remaining elements remained
unchanged, compared to the previous version. Figure 18 shows
the structure of the legs aer introducing the above changes.
Fig. 18. Geometry of the support structure aer optimisation.
Figures 19 and 20 show the stresses and deformations
of the nal version of the support structure. Based on their
analysis we can conclude that the highest stresses, exceeding
400 MPa, only occur in single nite elements of the model, in
the connection areas of structure elements. It is noteworthy
that the calculations making use of 2D coating elements can
reveal unrealistically increased stresses in the areas of rapid
geometry changes, therefore the authors allowed the critical
stresses to be slightly exceeded in the one-element band from
the edge connecting two elements of the structure.
In order to obtain more reliable results in the contacting
areas of main elements of the structure, additional zonal
analyses are to be performed with the use of 3D elements for
selected areas.
e strength of the structure in the above described areas
will be ensured for the assumed load distribution by the use
of high-strength steel.
e mass of the structure aer the introduced changes
concerning stieners and sheet thickness is equal to 1582 t.
SELECTED ECONOMIC ASPECTS
For the nal support structure, additional parameters were
calculated which can be of certain importance for investors as
aecting the predicted investment costs. As already mentioned,
the mass of the steel structure is equal to about 1582 t, without
welds and preservation coatings. e information on the mass
of the foundation and the unit cost of production of 1 kg of
ship steel structures, which at present amounts to about 4 €,
enables to assess the cost of production of the steel structure
alone as approximately equal to 6.328.000 €. Based on the
structure geometry analysis, the volume of the inside of the
Unauthenticated
Download Date | 3/12/19 12:47 PM
POLISH MARITIME RESEARCH, No 1/2015
45
foundation to the predicted level of ballast lling was also
calculated as approximately equal to 9 thousand m3. Taking
into account that the role of ballast will be played by the pulp
in the form of wet gravel having the unit mass equal to 1700
kg/m3, the mass of the entire ballast will equal 15300 t, which
at the current unit price equal to about 25 PLN (6€) for 1 m3
of the pulp gives 91800 €.
Fig.19. Reduced stresses [MPa]
Fig. 20. Elastic deformations [mm]
SUMMARY
e art icle refers to current, a nd simultaneously very complex
and important issues connected with designing marine objects
to support large wind power plants. Authors made an attempt
to design a steel support structure of a tripod type which is
intended to be xed on the Baltic seabed, at a depth of about
60 m. Structures of this type which were built in the past were
installed at smaller depths and linked with the seabed using
piles. e here presented structure is of gravitational type,
hence a number of dicult and completely new problems were
to be solved. It was done by successive modications of the
designed structure with respect to its both geometrical and
strength parameters, and to work out numerical models and
perform calculations with the aid of advanced numerical codes.
e presented nal version of the structure meets the assumed
requirements. e mass of the structure is comparable with
the mass of another simultaneously designed steel structure,
also of gravitational type, which is intended to be founded at
the depth of 40 m.
It is noteworthy that over 85 % of the total mass of the
structure is planned do be made of normal strength hull
structural steel (class NV B – Re 235, according to [8]). is
makes this project more advantageous and attracting strong
interest of representatives of wind farm investors. In most
heavily loaded areas of the structure the steel NV AEH420 –
Re 420 was used, according to [8].
In authors’ opinion, the article names and discusses
a number of issues concerning the subject matters of designing
of steel marine structures which will be highly applicable
for designers, research workers, and investors interested in
those types of objects. Moreover, it enables to evaluate the
level of technological and material costs connected with the
production of the examined object.
ACKNOWLEDGMENT
FEM analysis has been performed with HYPER WORKS
soware. Calculations were carried out at the Academic
Computer Centre in Gdansk (TASK).
is research was supported by e Polish National Centre
for Research and Development (NCBR) under the project PBS1/
A6/8/2012 “AQUILO”
REFERENCES
1. Sarpkaya T.: Wave forces on oshore structures, Cambridge
University Press, 2010
2. Sarpkaya T.: In-line and transverse forces on smooth
and rough cylinders in oscillatory ow at high Reynolds
numbers, Monterey, California. Naval Postgraduate
School, 1986
3. Recommended Practice DNV-RP-C205: Environmental
conditions and environmental loads, Det Norske Veritas,
October 2010
Unauthenticated
Download Date | 3/12/19 12:47 PM
POLISH MARITIME RESEARCH, No 1/201546
4. Levis E.V.: Principles of Naval Architecture. Vol. III –
Motions in Waves and Controllability, SNAME, 1989
5. Dymarski P., Ciba E., Marcinkowski T.: Eective method
for determining environmental loads on supporting
structures for oshore wind turbines, 20th International
Conference on Hydrodynamics in Ship Design and
Operation HYDRONAV 2014, Wroclaw, Poland, June 2014
6. Turbine graphics, source: Repower (in Polish)
7. PN-80/B-03040 „Foundations and supporting structures
for machines” (in Polish)
8. DNV-OS-J101 Design of Oshore Wind Turbine Structures
CONTACT WITH THE AUTOR
Czesław Dymarski,
Paweł Dymarski,
Jędrzej Żywicki
Gdańsk University of Technology
Faculty of Ocean Engineering and Ship Technology
11/12 Narutowicza Str.
80-233 Gdańsk
POLAND
Unauthenticated
Download Date | 3/12/19 12:47 PM
