Waves and wave forces:

In this task some considerations on the wave and current conditions and their associated loads on a
circular bridge support will be addressed. A bridge is planned across the sound between the mainland
A and an island B. To support the bridge a 13 m diameter circular caisson is to be placed midway in
the crossing. A flat sea bottom at the location of the caisson is built up of gravel and sand of various
grain diameters. The water depth at the caisson is to be 26 meters. The design waves are coming from
a sector 315 -345 and there is a semidiurnal tidal current of 1.0 m/s amplitude as indicated by the
double-ended arrow on the sketch. In an earlier project wave statistics was provided at an open water
location C, North-West of the site where the water depth was 150 m. To determine wave loads on the
structure it is found appropriate to use those data, but using only data for the waves coming from the
sector 315 -345. Tables for joint occurrences of sea state parameters are provided on page 6 and 7.

1) For a wave period of 7.0 s determine the lowest wave height that will move sea bed sand at the
depth of 26 m if a near bottom water particle velocity of 30 cm/s is required to initiate motion.

After calculating the wavelength using the approx. formula we find that we need to apply the general
sets of wave theory formulas given for intermediate depths. Be aware of the possible error using the
approximative formula for wave length. It is the argument for the tanh() that should be raised in the
power of and not (tanh()). The approx wavelength becomes 73.58 m. Thereafter by using formula
cosh
sin
sinh
k(z+h)
u = a ( t - kx)
kh
e e we obtain
2 2
1 7 2
sinh sinh 0.3sinh 26 1.5
73.6
cr cr cr
a = u u
T
kh kh m
e t t
t
= = =

and that
3.0
cr
H m =
.

2) In a sea-state described by a Rayleigh distribution and significant wave height equal to 3.0 m and
mean wave period equal to 7.0 s,
A
B
waves
C
a) what will be the probability that an arbitrary individual wave exceeds 3.0 m in height ?
Using the formula for the Rayleigh distribution and remember that in its formpresented gives
the probability that the wave height is less than a given value. Ssince we want the probability
of exceeding a certain value this is given by 1 F(h) and for the given values it becomes
0.135.

b) what would be the expected highest wave in this sea-state assuming the duration of the sea
state is 2 hours?
Here we use the formula for Hmax. N is the number of waves which is easily found by using
the information given by the mean wave period. It is defined as the total time divided by
number of waves so N = 2 hrs/7 s = 1028 waves. Result is Hmax = 5.8 m.

c) what wave period would you recommend to use for wave kinematics calculation for the
expected highest wave you determined in 1.b)? J ustify your answer.
Different answers were validated, however no score if you suggested a shorter period,
Information of figure 1 (cf home assignment) was the best source in this examset, indicating
that the wave period for the higher waves in sea state is approx 1.2 times the mean wave
period, so I would say that T = 8.4 s was the best answer.



3) A long-term statistical analysis is required to determine the design wave conditions at location C.

a) Use the joint occurrence table H
m0
-Mean Direction for the sector 315 -345 to establish a
table for plotting the observed distribution of significant wave height on the attached Weibull-
scaled paper. Use H
m0
intervals of 0.4 m. Use the table to estimate the remaining parameters of
the 3-parameter Weibull distribution that fits the data in a best way by using the Weibull
plotting paper. Assume H
0
=0 in the 3-parameter Weibull distribution. Include the plotting
table in your answer.
( )
0
0 0
0
Pr ( ) 1 exp ,
m
c
h H
H h F h h H
H H

| |
| |
| s = = >
|
|

\ .
\ .

See enclosed diagram.

b) The data in the jointoccurrence tables are based on data sampled every 3 hours over 1490
days. Calculate the average time between the observations within the sector 315 -345 and
determine the appropriate probability levels to be used to determine the 10 and 30 year
significant wave height at location C.

Given:
0,
1 ( )
p
p
m R
R
F H
t
=

Average time between observations: 1490 days/1246 observations = 1.1958 days = 28.7 h.

0,
( ) 1
p
m R
p
F H
R
t
= where 28.7h t = results in
10 yr:
0,10
28.7
( ) 1 0.99967
10 365 24
m
F H = =


30 yr:
0,10
28.7
( ) 1 0.99989
30 365 24
m
F H = =



c) Determine the 10 and 30 year significant wave height at location C. (If you did not find a
solution in 3.b) use =1.5, Hc =1.2 m and H
0
=0 m.)
( ) ( ) ( )
1/
0
0 0
0
( ) 1 exp ln 1 ( )
c
c
h H
F h h H H H F h
H H

| |
| |
| = = +
|
|

\ .
\ .

10 yr: ( ) ( )
1/1.5
0,10
1.2 ln 1 0.99967 4.8
m
h H m m = = =
30 yr: ( ) ( )
1/1.5
0,10
1.2 ln 1 0.99989 5.2
m
h H m m = = =

4) Wave and current forces must also be considered. Assume that the design wave condition is
H =3.0 m and T =7.0 s and the design current speed is 1.0 m/s. Assume that the caisson is smooth.

a) Determine what set of formulas you would apply to estimate the total horizontal force on the
caisson in the case of waves only. Give reasoning for your choice.
The Keulegan Carpenter number is essential. K = u
0
T/D. Wave length as in problem3.1:
73.5 m. This gives for the horizontal particle velocity at the surface:
0
0
cosh( ( )
1 2 1
1.5 1.38 /
2
sinh( ) tanh( ) 7
tanh( 26)
73.5
z
k z h
u a a m m s
kh kh s
t
e e
t
=
+
= = = =
and 1.38 7/13 0.74 K = = Since K is less than 2 potential theory must be used. With a
vertical circular cylinder ranging fromthe sea bed and through the sea surface MacCamy and
Fuchs te\ory is well suited. Alternatively the inertia termof Morison formula for force per
unit cylinder axis length can be applied:
2
( 0, , )
4
x M M
D
f f C u x z t t = = = with a C
M
dependent on the diameter to wavelength ratio
and
2
cosh ( )
cos( )
sinh
k z h
u a t
kh
e e
+
= fromlinear wave theory. C
M
=2.0 would be a good
choice here since D/L=0.18.

b) Calculate the total horizontal wave induced force in the direction of the waves and the
associated overturning moment about the sea bed. At least put up the formal expressions to
calculate the forces and the overturning moment (half score).

Using f
M
:

0 2
2
0
2
2
2
0
2
2 2
2 2
2
cosh ( )
cos( )
4 sinh
1
cos( ) cosh ( )
4 sinh
1 1
cos( ) sinh ( )
4 sinh
1 4 73.5
cos( ) 1025 2 6.5 1.5 3846.7
4 7 2
X M
h
M
h
M
h
M
D k z h
F C a t dz
kh
D
C a t k z h dz
kh
D
C a t k z h
kh k
D
C a t kN
k
t e e
t e e
t e e
t
t e e t
t

+
=
= +
= +
= = =
}
}

( )
0 2
2
cosh ( )
cos( )
4 sinh
M
d
D k z h
M C a t d z dz
kh
t e e

+
= +
}

Consider:
( )
( ) ( )
0 0
0 0
0 0
2 2
2 2
cosh( ) cosh cosh sinh sinh
cosh cosh sinh sinh
1 1
cosh cosh sinh sinh sinh cosh
1 1
cosh cosh sinh
d d
h h
h h
I kz kh zdz kz kh kz kh zdz
kh kz zdz kh kz zdz
z z
kh kz kz kh kz kz
k k k k
h
kh kh kh
k k k



= + = +
= +
| | | |
= + + +
| |
\ . \ .
| |
= +
|
\ .
} }
} }
( )
2
2 2
2 2 2
2
1
sinh sinh cosh
1 1 1
cosh cosh sinh cosh sinh sinh cosh
1
1 cosh
h
kh kh kz
k k
h h
kh kh kh kh kh kh kz
k k k k k
kh
k
( ( | | | |
+ +
| |
( (
\ . \ .
( ( | | | | | |
= + + +
| | |
( (
\ . \ . \ .
=

Where Rottmann gave solution for the integrals needed:
2
1
cosh( ) cosh sinh
x
x ax dx ax ax C
a a
= + +
}

2
1
sinh( ) sinh cosh
x
x ax dx ax ax C
a a
= + +
}


Insert this in the expression for M and solve:
( )
( )
0
2
2
0 2
2
0 0 2
2
2
2
2 2
cosh ( )
cos( )
4 sinh
1
cos( ) cosh ( )
4 sinh
1
cos( ) cosh ( ) cosh ( )
4 sinh
1 1 1 1
cos( ) sinh cosh
4 sinh
M
h
M
h
M
h h
M
D k z h
M C a t h z dz
kh
D
C a t k z h h z dz
kh
D
C a t k z h hdz k z h zdz
kh
D
C a t h kh
kh k k k
t e e
t e e
t e e
t e e

+
= +
= + +
| |
= + + +
|
\ .
= +
}
}
} }
2
2
1 sinh 1 cosh
cos( )
4 sinh
sinh cosh 1
3846.7kN 16.6m 63855kNm; 16.6m
sinh
M
x
kh
D kh kh kh
C a t h
k kh kh
kh kh kh
F R where R h
kh kh
t e e
| |
|
\ .
+
=
+
= = = = =
R is the moment arm( cf. with the same expression for MacCamy and Fuchs theory).

Using MacCamy and Fuchs theory:


2 2
2 2 1025 9.81 3 2 2
tanh( ) ( ) tanh 26 6.5 cos( )
73.5 73.5
2
73.5
x
gH
F kh A ka A t
k
t t
e o
t

| | | |
= =
| |
\ . \ .
| |
|
\ .

FromTable: A(0.5556)=?
By interpolation:

0.482 0.452 0.482
0.56 0.54 0.56 0.5556
0.482 1.5 0.0044 0.4754
A
A

=

= =


This gives: Fx = 3 834 kN
M=Fx R =3 834 16.6 =63643 kNm.

sinh cosh 1
16.6m
sinh
kh kh kh
where R h
kh kh
+
= =

The discrepancy between the values found using the two methods is negligible and a small
correction of C
M
would adjust for this.

c) Determine the period and the maxium value of the transversal force acting on the caisson
during the design current condition. Assume C
L
=0.8.
In steady flows the dominant transverse fluctuating force is caused by vortex shedding. The
vortex shedding frequency is determined by the Strouhal number which takes values around
0.2 see experimental results. The Strouhal number is given in the attachment as
0
f D
St
U
= . It depends on the Reynolds number Re
UD
v
= . Here Re =1300000. Fromthe
graph of experimental results St =0.26 so f
0
=0.26U/D=0.02 Hz
=>Period of oscillation T
0
=50 s.
The magnitude of this oscillating force is:
2 2
1
0.5 1025 0.8 13 26 1 138.6 kN
2
L L p
F C AU = = =


H
c
=0.9 m
H F
0.4 0.08741
0.8 0.373697
1.2 0.653569
1.6 0.835605
2.0 0.935044
2.4 0.973536
2.8 0.987971
3.2 0.994387
3.6 0.996792
4.0 0.999198
= 1.35
Comment: In my choice of fitted line I gave more attention
to the highest wave heights than the lowest, since we are
interested in the extremes. Even so, it is not easy to select
the best fit. Other lines may also be viewed as best fit. This
demonstrates that there are some uncertainties in estimating
the best fit. Maybe if another distribution function was used,
the data might fit better. In fact trying H
0
different from zero
could also help.
1