Proceedings of the ASME 2009 28th International Conference on Ocean, Offshore and Arctic Engineering

OMAE2009
May 31 - June 5, 2009, Honolulu, Hawaii, USA

Proceedings of the ASME 28th International Conference on Ocean, Offshore and Arctic
OMAE2009-80128
Engineering
OMAE2009
May 31-June 5, 2009, Honolulu, USA

OMAE2009-80128

EXPERIMENTAL VARIATION OF FOCUSING WAVE GROUPS FOR THE

INVESTIGATION OF THEIR PREDICTABILITY

Janou Hennig∗ Christian E. Schmittner

MARIN MARIN
Netherlands Netherlands
j.hennig@marin.nl c.schmittner@marin.nl

ABSTRACT ders [4]. [5] investigated capsizing of smaller vessels by genera-

In deterministic model testing, focusing wave groups are
used for the simulation of dedicated wave environments. They
are characterized by the transient appearance of one relatively
steep wave crest. The phasing of the wave components which
leads to an exact focusing in one point in time and space is
strongly dependent on the correct modeling of the wave phase
velocity while the position of the focusing point depends on the
wave group celerity. For wave generation purposes, the calcula-
tion of a wave maker control signal based on a target wave train
at a desired position in the tank (inverse or backward modeling)
is of crucial importance. Numerical wave tanks and empirical
approaches are often calibrated based on wave characteristics
measured in a particular tank. This paper presents model test
results for the variation of frequency range, steepness and fo-
cal point of focusing wave groups at intermediate water depth.
The measured characteristics are compared to predicted param-
eters.
INTRODUCTION
In deterministic model testing, focusing wave groups are
used for the simulation of dedicated wave environments. They
are characterized by the transient appearance of one relatively
steep wave crest and by well-defined periods and phase rela-
tions. They can therefore be used to model deterministic wave
sequences for model tests [1], [2], [3]. Thus, transient waves can
be used for the generation of breaking waves on vertical cylin-
∗ Address all correspondence to this author.
1 c 2009 by ASME
tion of large plunging breakers (freak waves). Wave packets are
good laboratory equivalents to single large ocean waves since
they are transient and show frequency focusing. Focused com-
ponent waves, even if still linear describable, behave in a fully
non-linear manner in a relatively small region around the con-
centration point (compare also [6]). They are also used to gener-
ate deterministic environments for the analysis of capsizing pro-
cesses [3], [7].
The phasing of the wave components which leads to an exact
focusing in one point in time and space is strongly dependent on
the correct modeling of the wave phase velocity while the posi-
tion of the focusing point depends on the wave group celerity. A
focusing wave group or transient wave packet consists of subse-
quently generated waves with increasing propagation speeds so
that all components meet in the so-called focusing or concentra-
tion point. After that point, they diverge in opposite order.
Transient waves for model excitation were originally pro-
posed by [8] and further developed by [9] and [10]. [11] recom-
mended a special type of transient waves called Gaussian wave
packets which have the advantage that their propagation behav-
ior can be predicted analytically [12]. With increasing efficiency
and capacity of computers the restriction to a Gaussian distri-
bution of wave amplitudes has been abandoned, and the entire
process is performed numerically [13]. Since they have a short
extension and consequently a small sensitivity with regard to dis-
turbing frequencies they can be used to determine the RAOs of
models within only one test run. The shape and width of the
wave spectrum can be selected individually for providing suffi-
Copyright °
 Table 1. OVERVIEW OF FOCUSED WAVE PACKETS DISCUSSED IN
cient energy in the relevant frequency range. [14] extended the
THIS PAPER.
linear approach to the generation and analysis of very high tran-
sient wave packets.
Wave No ω [rad/s] ζmax [m] Breaking
For wave generation purposes, the calculation of a wave
maker signal based on a target wave train at a desired position Wp 1 1 . . . 11 0.05 no
in the tank (inverse or backward modeling) is of crucial impor-
Wp 2 1 . . . 11 0.15 yes
tance. Numerical wave tanks and empirical approaches are often
calibrated based on wave characteristics measured in a particu- Wp 4 1...9 0.05 no
lar tank. This paper presents model test results for the variation
of frequency range, steepness and focal point of focusing wave Wp 5 1...9 0.15 yes
groups at intermediate water depth. The measured characteristics Wp 6 1...9 0.12 yes
are compared to predicted parameters.
Wp 7 4 . . . 11 0.05 ?
Wp 8 4 . . . 11 0.15 yes
TEST SETUP AND MODEL SCALE
Wave measurements were carried out at MARIN’s Shallow Wp 10 0.5 . . . 7 0.05 no
Water Basin which has the following dimensions: Wp 11 0.5 . . . 12 0.05 no
• Length: 220 m Wp 12 0.5 . . . 7 0.15 no
• Width: 15.8 m
• Water depth: 1.0 m Wp 13 0.5 . . . 7 0.25 yes
• Piston type wave generator Wp 14 0.5 . . . 12 0.15 yes
The following positions in terms of distances from the wave
board were selected for the wave probes:
• Close to the wave maker at x = 10 m sured at one of the subsequent positions in the wave tank:
• Between the latter position and the focal point at x = 20 m
and x = 40 m • Close to the wave maker at x = 10 m
• Area of focal point: x = 48, 50, 51, 52, 54, 56 m • Two intermediate points at x = 20 and 40 m
• Behind the focal point at x = 61 m • Target focal point at x = 49.50 m
• In the vicinity of the focal point at x = 50, 51, 52, 54 m
All dimensions are given in model scale throughout the pa- • Behind focal point at x = 61 m.
per. Thus, the values can be replaced by values scaled to desired
full scales or other tank scales. For better comparison, the time and wave elevation scale is kept
constant for all positions. Each picture contains both the mea-
sured wave train in red and the targeted wave train calculated
VARIATION OF WAVE CONDITIONS by linear wave theory (based on the wave train to be generated
The focusing wave trains studied in this paper are summa- by the wave maker in blue). There is hardly any difference be-
rized in Table 1. In our study, also further application exam- tween both time traces. In the first diagram, it even seems that
ples were included which will be described in a later section. there is only one graph present. This first diagram shows the still
The spectrum used for the focusing was optimized earlier for very small and long wave train with the shorter waves coming
the generation of focusing wave groups, in particular to avoid first, followed by the longer waves. On their way through the
wave breaking prior to the focal point. Basically, it could be re- wave tank, shorter waves are caught up by the longer and there-
placed by other arbitrary types of spectra. The number n of wave fore faster wave components such that they superimpose in one
frequencies was determined by the number of time samples N; focusing point at 49.50 m from the wave board.
n = N2 − 1 with a frequency step width of 4ω = n4t2π
, 4t sample If we zoom in at the position of the focusing point and
rate. 0.50 m further downstream, which is given in Fig. 2, we can
identify some slight differences between the measured and the
target wave train. However, we conclude that these differences
RESULTS are rather due to some general model test inaccuracy than due
A Really Linear Focusing Wave Group to non-linear wave propagation. As this wave group can be pre-
The focusing wave group Wp 1 serving as a reference case dicted by linear wave theory, even in the area of the focusing
is shown in Fig. 1. Each picture shows the wave time trace mea- point, the choice of WP 1 as a linear reference case is justified.

2 c 2009 by ASME
Copyright °
 0.05 x = 10 m
0 0.05
0
−0.05 −0.05
30 40 50 60 70 80 90 100 110 120 130 140 40 50 60 70 80 90 100 110 120
0.05 x = 20 m
0 0.05
0
−0.05 −0.05
30 40 50 60 70 80 90 100 110 120 130 140 40 50 60 70 80 90 100 110 120
0.05 x = 40 m
0 0.05
0
Wave elevation [m]

−0.05 −0.05

Wave elevation [m]

30 40 50 60 70 80 90 100 110 120 130 140 40 50 60 70 80 90 100 110 120
0.05 x = 49.50 m (linear focusing point)
0 0.05
0
−0.05 −0.05
30 40 50 60 70 80 90 100 110 120 130 140 40 50 60 70 80 90 100 110 120
0.05 x = 50 m
0 0.05
0
−0.05 −0.05
30 40 50 60 70 80 90 100 110 120 130 140 40 50 60 70 80 90 100 110 120
0.05 x = 56 m
0 0.05
0
−0.05 −0.05
30 40 50 60 70 80 90 100 110 120 130 140 40 50 60 70 80 90 100 110 120
0.05 x = 61 m
0 0.05
0
−0.05 −0.05
30 40 50 60 70 80 90 100 110 120 130 140 40 50 60 70 80 90 100 110 120
time [s] time [s]

Figure 1. LINEAR FOCUSING WAVE GROUP WP 1 MEASURED AT Figure 3. FOCUSING WAVE GROUP WP 4 MEASURED AT DIFFER-
DIFFERENT POSITIONS IN THE TANK (ω = 1...11 rad/s, ζmax = ENT POSITIONS (ω = 1...9 rad/s, ζmax = 0.05 m).
0.05 m).

0.05 ω = 1 . . . 9 rad/s) is shown. We can note that it is still predicted

Measured
Targeted
quite well by linear wave theory up to 10 m in front of the focus-
ing point. In the focusing point itself and 0.50 m further down-
Wave elevation [m]

0
stream the deviation becomes more obvious (Fig. 4). By looking
at the wave asymmetry, it can be seen that the focusing point
must be somewhere between 49.50 m and 50 m, thus slightly
further downstream than predicted by linear theory (In time do-
−0.05
110 111 112 113 114 115 116 117 118 119 120 main, a trough being deeper on the left hand side indicates a sit-
uation before focusing, a deeper trough on the right hand side
0.05 after focusing). A shift of the focusing point to another position
Measured
Targeted (without a change of phasing within the wave group itself) can be
Wave elevation [m]

observed for rather long (and higher) waves. It can be concluded

0 that this wave group seems to behave more ”low frequent” than
the reference case, and this only due to the fact that some higher
frequency components were not generated, but no lower frequen-
−0.05 cies added.
110 111 112 113 114 115 116 117 118 119 120
time [s] If we shift the frequency band even further to the low fre-
quent wave components, we see the following development of
Figure 2. LINEAR FOCUSING WAVE GROUP WP 1 CLOSE TO THE the wave group Wp 10 (Fig. 6). It seems that the wave train
FOCUSING POINT (ω = 1...11 rad/s, ζmax = 0.05 m). behaves still rather linear, but maybe a slight shift of the focus-
ing point to a location a bit further downstream can be suspected
(Fig. 7).
Variation of Frequency Bands (or Relative Water In Fig. 8, the focusing wave group Wp 7 with the same max-
Depth) imum wave elevation as our reference wave Wp 1, but reduced
Based on the reference case Wp 1, the frequency band is frequency band width (less low frequency wave components this
varied according to Table 1, resulting in the Fourier spectra com- time) is shown. In the focusing point itself and 0.50 m further up-
pared in Fig. 5. stream, we can state that the focusing is well established which
In Fig. 3, the focusing wave group Wp 4 with the same means that we see a very symmetric focusing shape of the wave
maximum wave elevation as our reference wave Wp 1, but re- in time. However, that wave train appears to arrive too early at
duced frequency band width (less high frequency components, the focusing point. Such, the same total maximum wave eleva-

3 c 2009 by ASME
Copyright °
 x = 49.50 m (linear focusing point) x = 10 m
0.05 0.05
Measured 0
Targeted −0.05
30 40 50 60 70 80 90 100
Wave elevation [m]

x = 20 m
0.05
0
0 −0.05
30 40 50 60 70 80 90 100
x = 40 m
0.05
0
−0.05

Wave elevation [m]

30 40 50 60 70 80 90 100
−0.05 x = 49.50 m (linear focusing point)
95 96 97 98 99 100 101 102 103 104 105 0.05
0
−0.05
x = 50 m 30 40 50 60 70 80 90 100
x = 50 m
0.05 0.05
Measured 0
Targeted −0.05
Wave elevation [m]

30 40 50 60 70 80 90 100
x = 56 m
0.05
0 0
−0.05
30 40 50 60 70 80 90 100
x = 61 m
0.05
0
−0.05
30 40 50 60 70 80 90 100
−0.05
95 96 97 98 99 100 101 102 103 104 105 time [s]
time [s]

Figure 6. FOCUSING WAVE GROUP WP 10 MEASURED AT DIFFER-

Figure 4. FOCUSING WAVE GROUP WP 4 CLOSE TO THE FOCUS-
ENT POSITIONS (ω = 0.5...7 rad/s, ζmax = 0.05 m).
ING POINT (ω = 1...9 rad/s, ζmax = 0.05 m).

x = 49.50 m (linear focusing point)

x = 10 m 0.05
1.2 Measured
WP 1 Targeted
Wave elevation [m]

WP 4
WP 7
WP 10
1 WP 11 0

0.8
−0.05
Spectral density [m s]

75 76 77 78 79 80 81 82 83 84 85
2

0.6 x = 50 m
0.05
Measured
Targeted
Wave elevation [m]

0.4

0.2

−0.05
75 76 77 78 79 80 81 82 83 84 85
0 time [s]
0 2 4 6 8 10 12 14
ω [rad/s]

Figure 7. FOCUSING WAVE GROUP WP 10 CLOSE TO THE FOCUS-

Figure 5. NORMALIZED SPECTRA OF ALL FOCUSING WAVE ING POINT (ω = 0.5...7 rad/s, ζmax = 0.05 m).
GROUPS WITH DIFFERENT FREQUENCY BANDS AT 10 m FROM
WAVE BOARD, WITH SAME MAXIMUM ELEVATION OF 0.05 m).
than theory and he low frequency components slightly faster than
linear theory would predict. It is still to be investigated why this
tion in the focusing point as in the previously discussed cases, is. As compared to lower frequency bands, the troughs around
combined with a higher frequency band, or removed low fre- the focal point are deeper. Also a shift of frequencies can be
quencies, results in an increase of phase speed. The focusing observed during the passing of the focal area if we compare the
appears to happen slightly earlier than predicted by linear the- spectra of WP 7 there (Fig. 9), possibly due to wave breaking.
ory. At x = 40 m, the phasing has become shifted, and it seems If we increase the frequency band of the linear focusing
that the high frequency components are travelling slightly slower wave group (reference case) at both sides, keeping the same max-

4 c 2009 by ASME
Copyright °
 x = 40 m x = 10 m
0.05 0.05
0
0 −0.05
40 50 60 70 80 90 100 110 120 130 140 150
−0.05 x = 20 m
90 95 100 105 110 115 120 125 130 0.05
x = 48 m 0
0.05 −0.05
40 50 60 70 80 90 100 110 120 130 140 150
0 x = 40 m
0.05
−0.05 0
90 95 100 105 110 115 120 125 130 −0.05

Wave elevation [m]

Wave elevation [m]

x = 49.50 m (linear focusing point) 40 50 60 70 80 90 100 110 120 130 140 150
0.05 x = 49.50 m (linear focusing point)
0.05
0 0
−0.05
40 50 60 70 80 90 100 110 120 130 140 150
−0.05 x = 50 m
90 95 100 105 110 115 120 125 130
x = 50 m 0.05
0.05 0
−0.05
40 50 60 70 80 90 100 110 120 130 140 150
0 x = 56 m
0.05
−0.05 0
90 95 100 105 110 115 120 125 130
x = 56 m −0.05
40 50 60 70 80 90 100 110 120 130 140 150
0.05 x = 61 m
0.05
0 0
−0.05
−0.05 40 50 60 70 80 90 100 110 120 130 140 150
90 95 100 105 110 115 120 125 130
time [s]
time [s]

Figure 8. FOCUSING WAVE GROUP WP 7 CLOSE TO THE FOCUS- Figure 10. FOCUSING WAVE GROUP WP 11 MEASURED AT DIFFER-
ING POINT (ω = 4...11 rad/s, ζmax = 0.05 m). ENT POSITIONS (ω = 0.5...12 rad/s, ζmax = 0.05 m).

x = 49.50 m (linear focusing point)

WP 7 0.05
1.2
Measured
x = 40 m
Targeted
x = 48 m
Wave elevation [m]

x = 50 m

1
0

0.8
Spectral density [m s]

−0.05
2

115 116 117 118 119 120 121 122 123 124 125

0.6 x = 50 m
0.05
Measured
Targeted
Wave elevation [m]

0.4

0.2

−0.05
115 116 117 118 119 120 121 122 123 124 125
0 time [s]
0 2 4 6 8 10 12 14
ω [rad/s]

Figure 11. FOCUSING WAVE GROUP WP 11 CLOSE TO THE FOCUS-

Figure 9. COMPARISON OF NORMALIZED SPECTRA OF FOCUSING ING POINT (ω = 0.5...12 rad/s, ζmax = 0.05 m). The focusing occurs
WAVE GROUP WP 7 AT DIFFERENT LOCATIONS AND AFTER WAVE further downstream than predicted by linear theory.
BREAKING).

components seem to focus at the same location and time: There

imum wave elevation in the focusing point, the wave packet (Wp are some higher frequency oscillations before and after the ac-
11) shows a relatively linear behavior in front of and behind the tual focusing. In this case, we can identify the phenomenon of
focusing point (Fig. 10, x = 10 m through x = 40 m and x = 56 m a shift in space of the focusing point as well as ”non-focusing”
through x = 61 m). However, zooming in close to the focusing of some components which could be described as some shallow
point reveals that the focusing occurs at a position further down- water effect combined with a result of a change in phasing due
stream than predicted by linear theory. Also, not all of the wave to non-linear wave-wave interaction.

5 c 2009 by ASME
Copyright °
 Summarizing our tentative observations, we can find the fol- x = 10 m
0.1
lowing: 0
−0.1
30 40 50 60 70 80 90 100 110 120 130 140
x = 20 m
• There seems to be a kind of optimum parameter set of wave 0.1
0
frequency band, maximum wave elevation (in the focus- −0.1
30 40 50 60 70 80 90 100 110 120 130 140
ing point) and water depth which characterizes the focusing x = 40 m
0.1
wave group as a linearly predictable one. This might be con- 0
−0.1

Wave elevation [m]

sidered a naive statement but will be necessary to develop a 30 40 50 60 70 80 90 100
x = 49.50 m (linear focusing point)
110 120 130 140

rule of thumb for deterministic wave generation. 0.1

0
−0.1
• Reducing the number of higher frequencies only, the wave 30 40 50 60 70 80 90
x = 50 m
100 110 120 130 140

train gets more of a shallow water wave character, its phase 0.1
0
speed gets closer to its group speed, and the frequency −0.1
30 40 50 60 70 80 90 100 110 120 130 140
x = 56 m
components propagate more or less dependent on the water 0.1
0
depth although the higher frequencies travel at their ”nor- −0.1
30 40 50 60 70 80 90 100 110 120 130 140
mal” speed relatively to the ”carrier wave”. Thus the fo- x = 61 m
0.1
cusing point occurs further downstream because from a cer- 0
−0.1

tain moment on the higher frequent waves are carried by 30 40 50 60 70 80

time [s]
90 100 110 120 130 140

the long shallow water waves which seems to increase the

group speed without affecting the phasing during this pro-
Figure 12. FOCUSING WAVE GROUP WP 2 WITH THREE TIMES
cess. Thus, the focusing point is further downstream, main-
HIGHER WAVE ELEVATION AT WAVE MAKER AS COMPARED TO WP
taining a symmetrical wave shape in time.
1 (ω = 1...11 rad/s, ζmax = 0.15 m).
• Considering a frequency band at higher frequencies only
leads to a focusing point earlier in time.
• Increasing the frequency band width in total, a combination x = 49.50 m (linear focusing point)
0.15
of the above phenomena occurs. Measured
0.1 Targeted
Wave elevation [m]

0.05

Increase of Initial Wave Elevation 0

Up to now, the variation of frequency band was considered −0.05

while keeping the wave elevation roughly constant. In the fol- −0.1

lowing, the wave elevation is increased systematically. In Fig. 12,

110 111 112 113 114 115 116 117 118 119 120
the focusing wave group Wp 1 which could be predicted by linear
wave theory (Fig. 1) is increased to a wave elevation of 0.15 m x = 50 m

instead of 0.05 m. This gives Wp 2 which first appears to be- 0.15

Measured

have linear, even up to 40 m distance from the wave maker. 0.1 Targeted
Wave elevation [m]

Somewhere between the wave probe at 40 m and the focusing 0.05

point, wave breaking occurs resulting in a somewhat deformed 0

time trace at the focal point. Shortly behind the focusing point, −0.05

the steepness increases dramatically once again (Fig. 13). Wave −0.1

breaking is likely to occur once more. Shortly further down- 110 111 112 113 114 115 116 117 118 119 120
stream, from x = 56 m on, it becomes evident that higher fre- time [s]

quency components are broken out of the wave train. This is also
confirmed by a comparison between the Fourier spectra of the
original wave and the increased wave (Fig. 21 on the last page).
Increasing the wave height of the wave group Wp 4 (reduced
frequency band at its higher end) to the same value of 0.15 m,
the wave Wp 5 does not behave as non-linear as the one with
the full frequency band (Fig. 14). Close to the focusing point,
it is evident that wave breaking has occurred (compare spectra
in Fig.15). However, the wave in Fig. 14 does not appear as
”collapsed” as Wp 4 in Fig. 3. The biggest evidence of non-linear
effects is the wave elevation which is much higher than predicted
by linear theory, even before the focusing point. Furthermore, the
Figure 13. FOCUSING WAVE GROUP WP 2 WITH THREE TIMES
HIGHER WAVE ELEVATION AT WAVE MAKER AS COMPARED TO WP
1 (ω = 1...11 rad/s, ζmax = 0.15 m).
actual focusing point is shifted to a more downstream position as
compared to the case with the smaller elevation. Thus, also non-
linear shallow water effects gain significance.
The focusing wave group Wp 10 with the lowest frequency
band from Fig. 7 shows as Wp 12 similar effects when increased
in elevation, but not as pronounced (Fig. 16). However, also here,

6 c 2009 by ASME
Copyright °
 x = 49.50 m (linear focusing point) x = 50 m
0.15 1.2
Measured WP5
0.1 Targeted WP 4
Wave elevation [m]

0.05
1
0

−0.05

−0.1 0.8

Spectral density [m s]
2
90 95 100 105

0.6
x = 50 m
0.15
Measured
0.1 Targeted
Wave elevation [m]

0.4
0.05

−0.05 0.2

−0.1

90 95 100 105 0
time [s] 0 2 4 6 8 10 12 14
ω [rad/s]

Figure 14. FOCUSING WAVE GROUP WP 5 WITH THREE TIMES

Figure 15. COMPARISON OF NORMALIZED SPECTRA OF FOCUS-
HIGHER WAVE ELEVATION AT WAVE MAKER AS COMPARED TO WP
ING WAVE GROUPS WP 4 AND WP 5 SHOWN IN FIG. 3 AND 14
4 (ω = 1...9 rad/s, ζmax = 0.15 m). Shortly before the focal point, the
(ω = 1...9 rad/s, ζmax = 0.05 m AND 0.15 m, RESPECTIVELY).
wave becomes significantly steeper and shows more vertical asymmetry
than predicted by linear theory. Furthermore, the focal point is shifted to
a position further downstream.
x = 49.50 m (linear focusing point)
0.15
Measured
0.1 Targeted
Wave elevation [m]

the occurrence of the focusing point is shifted even further down- 0.05

stream than compared to the lower case. 0

In Fig. 17, the wave group Wp 7 with the high frequency −0.05

band is increased in height which results in Wp 8 with massive −0.1

wave breaking already far away from the focusing point such that
110 111 112 113 114 115 116 117 118 119 120
no actual focusing is realized. The wave steepness at 40 m is ob-
viously increased and the high frequency components run ahead x = 50 m
0.15
the linear prediction. However, it seems that much further down- Measured
0.1 Targeted
stream the wave tries to focus with the remaining frequencies.
Wave elevation [m]

0.05

0
A closer look to the Fourier spectra in Fig. 18 reveals that
−0.05
the width of the spectrum is reduced the more the further down-
stream we look. −0.1

The wave packet with the largest frequency band, Wp 14, 110 111 112 113 114 115
time [s]
116 117 118 119 120

occurs very similar to the increased reference case Wp 2. Even

close to the focal point (Fig. 19), the time profile is very much
alike, except for steeper crests due to more high frequency com- Figure 16. FOCUSING WAVE GROUP WP 12 WITH THREE TIMES
ponents. HIGHER WAVE ELEVATION AT WAVE MAKER AS COMPARED TO WP
10 (ω = 0.5...7 rad/s, ζmax = 0.15 m).
We increased the maximum wave elevation further to
0.25 m, but due to limited space we just want to mention that
interestingly the general shape of wave group Wp 6 at 0.25 m
(Fig. 22) looks very much alike the wave group Wp 2 at 0.15 m
close to the focusing point. Furthermore, the wave trains tend to all components focus in one point in time and space anymore as
become more steepened and asymmetric (Fig. 23). Making the the higher frequencies propagate at higher speed as compared to
wave trains steeper, the focusing might be affected such that not linear wave theory.

7 c 2009 by ASME
Copyright °
 x = 40 m x = 50 m x = 61 m
1.2 1.2 1.2
WP 1 WP 1 WP 1
WP 2 WP 2 WP 2

1 1 1

0.8 0.8 0.8

Spectral density [m2s]

Spectral density [m s]

Spectral density [m2s]

2
0.6 0.6 0.6

0.4 0.4 0.4

0.2 0.2 0.2

0 0 0
0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14
ω [rad/s] ω [rad/s] ω [rad/s]

Figure 21. COMPARISON OF NORMALIZED SPECTRA OF FOCUSING WAVE GROUPS WP 1 AND WP 2 SHOWN IN FIG. 1 AND 12 (ω =
1...11 rad/s, ζmax = 0.05 m AND 0.15 m, RESPECTIVELY). ON ITS WAY THROUGH THE TANK, THE WAVE CRESTS STEEPEN DUE TO NON-
LINEAR WAVE-WAVE INTERACTION AND WAVE BREAKING OCCURS.

x = 10 m WP 8
1.2
0.1
0 x = 20 m
−0.1 x = 50 m
30 40 50 60 70 80 90 100 110 120 130 140 x = 61 m
x = 20 m
0.1 1
0
−0.1
30 40 50 60 70 80 90 100 110 120 130 140
x = 40 m
0.1
0 0.8
−0.1
Wave elevation [m]

Spectral density [m s]

30 40 50 60 70 80 90 100 110 120 130 140

2

x = 49.50 m (linear focusing point)

0.1
0
−0.1 0.6
30 40 50 60 70 80 90 100 110 120 130 140
x = 50 m
0.1
0
−0.1
30 40 50 60 70 80 90 100 110 120 130 140 0.4
x = 56 m
0.1
0
−0.1
30 40 50 60 70 80 90 100 110 120 130 140 0.2
x = 61 m
0.1
0
−0.1
30 40 50 60 70 80 90 100 110 120 130 140
time [s] 0
0 2 4 6 8 10 12 14
ω [rad/s]

Figure 17. FOCUSING WAVE GROUP WP 8 WITH THREE TIMES

HIGHER WAVE ELEVATION AT WAVE MAKER AS COMPARED TO wp Figure 18. COMPARISON OF NORMALIZED SPECTRA OF WAVE
7 (ω = 4...11 rad/s, ζmax = 0.15 m). GROUP WP 8 AT DIFFERENT LOCATIONS.

• Location of focusing shifted

CONCLUSIONS AND PERSPECTIVES • Wave breaking
The observed characteristics of the measured focused wave • Sharp and asymmetric crests
groups are summarized in Table 2. The non-linear behavior of • Change of frequency band
the wave trains can be characterized by the following phenom-
ena: An increase of crest height as well as a change to a sharpened
crest profile could be observed for Wp 5 (Fig. 14) and Wp 13
• Asymmetry in front of and behind focusing point (Fig. 23). Wp 12 showed an increase in crest height without
• Asymmetry in time of focusing wave group at location of asymmetry occurrence (Fig. 16).
focusing In general, it can be observed that a restriction in frequency
• Increase of absolute wave elevation in terms of the linear range leads to a better agreement of the actual ave propagation
prediction with linear theory. This could be due to reduced wave celerity
• Time of focusing shifted differences and less wave-wave interactions.

8 c 2009 by ASME
Copyright °
 x = 49.50 m (linear focusing point) x = 49.50 m (linear focusing point)
0.15
Measured 0.2 WP 2 (scaled up)
0.1 Targeted WP 6
Wave elevation [m]

Wave elevation [m]

0.05 0.1

0 0

−0.05
−0.1
−0.1
−0.2

150 151 152 153 154 155 156 157 158 159 160 90 95 100 105

x = 50 m x = 50 m
0.15
Measured 0.2 WP 2 (scaled up)
0.1 Targeted WP 6
Wave elevation [m]

Wave elevation [m]

0.05 0.1

0 0

−0.05
−0.1
−0.1
−0.2
150 151 152 153 154 155 156 157 158 159 160 90 95 100 105
time [s] time [s]

Figure 19. FOCUSING WAVE GROUP WP 14 WITH THREE TIMES Figure 22. FOCUSING WAVE GROUPS WP 2 AND WP 6 CLOSE TO
HIGHER WAVE ELEVATION AT WAVE MAKER AS COMPARED TO WP FOCUSING POINT.
11 (ω = 0.5...12.6 rad/s, ζmax = 0.15 m).

x = 40 m x = 61 m
1.2 1.2
WP 11 WP 11
WP 14 WP 14
x = 49.50 m (linear focusing point)
1 1

0.2 Measured
0.8 0.8 Targeted
Wave elevation [m]
Spectral density [m s]

Spectral density [m2s]

2

0.1
0.6 0.6

0
0.4 0.4

0.2 0.2
−0.1

0 0 −0.2
0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14
ω [rad/s] ω [rad/s]

110 111 112 113 114 115 116 117 118 119 120

Figure 20. COMPARISON OF NORMALIZED SPECTRA OF FOCUS- x = 50 m

ING WAVE GROUP WP 11 AND WP 14 at x = 40 m AND x = 61 m). 0.2 Measured

Targeted
Wave elevation [m]

0.1

As an example for the application of the wave packet tech- 0

nique, Fig. 24 shows a tailor-made high wave sequence embed- −0.1

ded in a sea state which is basically a focusing wave group gen- −0.2
erated deterministically in a realistic sea environment. This illus- 110 111 112 113 114 115 116 117 118 119 120
trate that although this technique is based on linear wave theory time [s]

it can be used to generate high wave events.

As the wave focusing technique can be used for calibrat- Figure 23. FOCUSING WAVE GROUP 13 WITH HIGHEST WAVE ELE-
ing wave tanks (for deterministic wave generation), we will gain VATION AT WAVE MAKER (WP 13, ω = 0.5...7 rad/s, ζmax = 0.25 m).
more insight in the opportunities that our wave tanks offer. Un-
derstanding of wave packets in model testing means understand-
ing wave theory and facilities as well as some major characteris-
tics and wave effects in wave hydrodynamics. In this sense, this
study might also serve as a case study in basic wave understand-
ing. ceived from this study. Furthermore, third order effects on com-
For the near future we plan to derive empirical formulas for ponents phase velocity could be used to derive the wave maker
deterministic wave generation based on the entire data set we re- motion and compared to the experiments discussed.

9 c 2009 by ASME
Copyright °
Table 2. OBSERVED CHARACTERISTICS OF MEASURED FOCUSED x = 40 m
WAVE GROUPS. 10
0
−10
Wave No Focusing Focus point [m] Time shift 2500 2550 2600 2650 2700 2750 2800 2850 2900 2950
x = 48 m (focusing point)

Wp 1 yes 49.50 small 10

0

Wp 2 no - - −10
2500 2550 2600 2650 2700 2750 2800 2850 2900 2950
x = 50 m
Wp 4 yes ≈ 49.75 small 10
0
Wp 5 asymmetric 50 small −10
2500 2550 2600 2650 2700 2750 2800 2850 2900 2950

Wave elevation [m]

x = 51 m
Wp 6 breaking 10
0
Wp 7 yes 49.50 significant −10
2500 2550 2600 2650 2700 2750 2800 2850 2900 2950
Wp 8 breaking - - x = 52 m
10

Wp 10 yes ≈ 49.60 small 0

−10
2500 2550 2600 2650 2700 2750 2800 2850 2900 2950
Wp 11 almost ≈ 49.80 small time [s]

x = 48 m (focusing point)
Wp 12 almost > 50 small 15

Wp 13 asymmetric > 50 -
10
Wp 14 breaking > 50 significant

5
Wave elevation [m]

REFERENCES 0
[1] Clauss, G., and Saroukh, A., 1996. “Interaction between
vessel, stinger, and pipeline during laying operations in
high seas”. In Proceedings of the 15 th Int. Conference on −5

Offshore Mechanics and Arctic Engineering (OMAE 96-

1027).
−10
[2] Clauss, G. F., and Kühnlein, W. L., 1997. “Simulation of
design storm wave conditions with tailored wave groups”.
In Proceedings of the 7th International Offshore and Polar −15
2760 2780 2800 2820 2840 2860 2880
Engineering Conference (ISOPE), pp. 228–237. time [s]

[3] Hennig, J., 2005. “Generation and Analysis of Harsh

Wave Environments”. Dissertation. Technische Universität
Berlin (D 83).
[4] Wienke, J., Sparboom, U., and Oumeraci, H., 2001. “Large
Scale Experiments with Slender Cylinders in Breaking
Waves”. In ISOPE 2001 - 11th International Offshore and
Polar Engineering Conference. ISBN 1-880653-54-0.
[5] Kjeldsen, S. P., 1983. 2- and 3-dimensional deterministic
freak waves. Tech. rep., Norwegian Hydrodynamic Labo-
ratories, Trondheim, November.
[6] Chaplin, J. R., 1996. “On Frequency-Focusing Unidirec-
tional Waves”. International Journal of Offshore and Polar
Engineering, 6(2), pp. 131–137.
[7] Clauss, G. F., Hennig, J., Cramer, H., and Brink, K.-E.,
2005. “Validation of Numerical Motion Simulations by Di-
rect Comparison with Time Series from Ship Model Tests
in Deterministic Wave Sequences”. In OMAE 2005 - 24th
Figure 24. DETERMINISTIC HIGH WAVE SEQUENCE EMBEDDED IN
SEA STATE AND ZOOMED IN AT FOCUSING POINT.
International Conference on Offshore Mechanics and Arc-
tic Engineering. OMAE2005-67123.
[8] Davis, M. C., and Zarnick, E. E., 1964. “Testing ship mod-
els in transient waves”. In 5th Symposium on Naval Hy-
drodynamics.
[9] Takezawa, S., and Hirayama, T., 1976. “Advanced experi-
mental techniques for testing ship models in transient wa-
ter waves. Part II: The controlled transient water waves for
using in ship motion tests”. In Proceedings of the 11th
Symposium on Naval Hydrodynamics: Unsteady Hydro-
dynamics of Marine Vehicles, R. Bishop, A. Parkinson, and
W. Price, eds., pp. 37–54.

10 c 2009 by ASME
Copyright °
[10] Mansard, E. P. D., and Funke, E. R., 1982. “A new approach
to transient wave generation”. In Proceedings of Coastal
Engineering.
[11] Bergmann, J., and Schmitz, R., 1985. Nobiskrug —
ölabschöpfkatamaran Erprobung in der nordsee im Septem-
ber/Oktober 1985. Tech. Rep. Bericht-Nr. TUB/ISM 85/11,
Institut für Schiffs- und Meerestechnik, TU Berlin.
[12] Chakrabarti, S. K., and Libby, A. R., 1988. “Further ver-
ification of Gaussian wave packets”. Applied Ocean Re-
search, 10(2).
[13] Clauss, G., and Kühnlein, W., 1995. “A New Approach
to Seakeeping Tests of Self-Propelled Models in Oblique
Waves with Transient Wave Packets”. In OMAE 95 - 350.
[14] Clauss, G. F., and Kühnlein, W. L., 1997. “A new tool for
seakeeping tests – nonlinear transient wave packets”. In
Proceedings of the 8 th Int. Conference on the Behaviour of
Offshore Structures (BOSS), pp. 269–285.

11 c 2009 by ASME
Copyright °