Environmental Engineering Reference
In-Depth Information
Up to 10
9
load cycles for wind loads alone can be expected over the same period [23].
The sea state is customarily described on the basis of short- and long-term statistics.
The methods used for this are described in detail in Section 2.5.
4.9.2
Fatigue analyses according to DIBt wind turbine guideline
4.9.2.1 Simpli
ed analyses for concrete
According to [9], a more accurate analysis of the concrete for repeated compressive
loads is unnecessary in wind turbine support structures with N
nom
2
10
9
load cycles
provided the condition according to Equation 4.1 is adhered to:
S
cd
;
max
0
:
40
þ
0
:
46
S
cd
;
min
(4.1)
Here, S
cd,min
and S
cd,max
denote the minimum and maximum effective concrete
compressive stress respectively due to the design load cases of Tables 4.1 or 4.4,
group F, to be investigated. They are calculated using Equations 4.2 and 4.3:
S
cd
;
min
¼ g
sd
s
c
;
min
h
c
=
f
cd
;
fat
(4.2)
S
cd;max
¼ g
sd
s
c
;max
h
c
=
f
cd;fat
(4.3)
where
g
Sd
¼
1.10 partial safety factor for modelling inaccuracies in the stress calculation
s
c,max
max. concrete compressive stress
s
c,min
min. concrete compressive stress at the same point at which
s
c,max
occurs,
calculated for the lower value of the action (use
s
c,min
¼
0 for tensile stresses)
h
c
factor for taking into account the non-uniform distribution of the concrete
compressive stresses according to Equation 4.4 (
h
c
¼
1.0 may be used for
simplicity)
The maximum concrete compressive stress at the extreme fibres may be reduced by the
factor
h
c
according to Equation 4.4 in the case of eccentric fatigue loads. This takes into
account the way the redistribution of stresses within the cross-section has a positive
influence on the resultant fatigue strength.
Therefore, the S-N curves derived from the uniaxial fatigue tests can be used when
determining the number of fatigue cycles to failure, also in the case of eccentric action
effects.
1
h
c
¼
(4.4)
5
s
c1
j
j
1
:
5
0
:
j
s
c2
j
Figure 4.27 shows the stress distribution for calculating
h
c
according to Model Code 90
[66]. Figure 4.28 shows this distribution transferred to a tower cross-section.
The reason for limiting the distance from the outer edge to x
300mm
t(t
¼
shaft
wall thickness) in Model Code 90 [66] is given in [49], which states that stress
redistributions are to be especially expected in structural cross-sections with a low
