Professional Documents
Culture Documents
ITTC - Recommended Procedures: Testing and Extrapolation Methods Loads and Responses, Sea Keeping Sea Keeping Experiments
ITTC - Recommended Procedures: Testing and Extrapolation Methods Loads and Responses, Sea Keeping Sea Keeping Experiments
5 – 02
07 - 02.1
Procedures Page 1 of 16
Testing and Extrapolation Methods
Effective Date Revision
Loads and Responses, Sea Keeping 2002 01
Sea Keeping Experiments
CONTENTS
1. PURPOSE OF PROCEDURE
2. SEAKEEPING EXPERIMENTS
2.1 Model Size
2.2 Model Completeness
2.3 Model Weight Distribution
2.4 Guidance System
2.5 Free Running Test
2.6 Measurement of Wave Loads
2.7 Measurement of Added Resistance
2.8 Measurement of Impact Load
2.9 Parameters to be Measured
2.10 Headings
2.11 Regular Waves
2.12 Transient Waves
2.13 Irregular Waves
2.14 Data Presentation
3. PARAMETERS
3.1 Parameters to be Taken into Account
3.2 Recommendations of ITTC For Parameters
4. VALIDATION
4.1 Uncertainty Analysis
4.4 Benchmark Tests
APPENDIX
A.1 EXAMPLE OF AN UNCERTAINTY ANALYSIS
A.1.1 Analysis of Elemental Errors
A.1.2 Response Amplitude Operator (RAO) and Overall Uncertainty
A.1.3 Results
A.1.4 Conclusions
Updated Approved
Seakeeping Experiments
1. PURPOSE OF PROCEDURE
To ensure that seakeeping tests to primarily 0.7
Fn . ω . √ ( Lm / g )
0.4
2. SEAKEEPING EXPERIMENTS
0.3
0.1
The size of the model should be such that
tank wall interference is avoided for the range 0
0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
of wave frequencies and model speeds to be BT / Lm
tested. Table 1, Fig.1 give, in dimensionless Fig.1 Maximum frequency at which tank inter-
form, a relationship among model length LM, ference occurs in head waves
tank breadth BT, Froude number Fr and the
highest wave frequency ω at which interfer-
ence effects may occur in head waves. 2.2 Model Completeness
need to be simulated. For tests in head or fol- 2.7 Measurement of Added Resistance
lowing waves with a model restrained in roll, it
is not necessary to simulate transverse weight None
distribution.
If the longitudinal radii of gyration for pitch 2.8 Measurement of Impact Load
or yaw are unknown, a value of 0.25 L pp
should be used. If the transverse radius of gyra- None
tion is unknown, a value between 0.35B and
0.40B, depending on the ship type, should be
used. (These values are those without including 2.9 Parameters to be Measured
the effect of added mass.)
The amplitudes and phases of hull motions
For experiments in which roll is not re- in desired degrees of freedom (as defined in
strained, the metacentric height should be Reference 6) should be measured as appropri-
simulated. If the vertical position of the centre ate to the purpose of the test.
of gravity is unknown, it should be determined
and recorded. As an alternative to ballasting the Wave height measurements should be made
model to a specified transverse radius of gyra- with a probe mounted close to the model, but
tion, the natural period of roll may be simu- not causing interference. The probe should
lated. preferably be fixed to the carriage, but meas-
urements may be made at a fixed point in the
tank. In the latter case, the measuring point
2.4 Guidance System should be selected in the position where waves
are fully formed without being affected by the
The guidance system should be such as to waves reflected at the wave maker and the tank
impose the minimum restraint on the motions walls & beaches.
of the model. It is desirable that in head or fol-
lowing waves the model should have the free- Establishments should develop the capabil-
dom to roll, that is, to rotate about the longitu- ity to measure the following additional parame-
dinal axis through the centre of gravity. In ters:
oblique waves, care must be taken on the re-
straint of sway and yaw motions. - Accelerations. In order to provide corrobo-
rating data for computation of accelera-
tions from measured motions.
2.5 Free Running Test
- Relative motion. Preferably measure-
None ments of the relative motion between the
model and the water surface at points that
2.6 Measurement of Wave Loads allow correlation with wave and other mo-
tion data.
None
ITTC – Recommended 7.5 – 02
07 - 02.1
Procedures Page 4 of 16
Testing and Extrapolation Methods
Effective Date Revision
Loads and Responses, Sea Keeping 2002 01
Sea Keeping Experiments
Tests should be carried out in waves appro- The coordinate system in which data are
priate to the sea conditions in which the vessel presented should be defined. Motion compo-
may be required to operate. In the absence of nents should also be defined. Linear transla-
specific wave spectrum data, an ITTC spec- tions and rotations may be presented in non-
trum should be used. In generating irregular dimensional forms as being divided by wave
waves in a tank, the input signal to the wave elevation and wave slope respectively. Rudder
maker should be produced in such a way that angles may be presented in other appropriate
the generated waves are non-repeatable. non-dimensional forms. Dimensional presenta-
tions can be more appropriate depending on the
Data should preferably be digitised before objectives of the experiment. Phase angles
analysis, using sample rates appropriate for the should be given in degrees and increases in
avoidance of aliasing with the individual meas- resistance and propulsion parameters should be
ured parameters. Care must be taken for the presented in nondimensional forms defined in
duration of the data acquisition so that enough Reference 6. Accelerations should be made
data are recorded for the objective of the test. non-dimensional by LPP/(g ς a ). It is recom-
The sample rate in the data acquisition needs to mended that the results are plotted to a base of
be fast enough in order that a sufficient resolu- ω (LPP/g)1/2 or ω e (LPP/g)1/2, although, depend-
tion is achieved.
ing on the objectives of the experiment, other
Energy spectra of waves and relevant re- bases such as wavelength-ship length ratio or
sponses should be produced through spectral wavelength may be appropriate. The limit of
analysis using either the indirect method of tank wall interference effects should be indi-
Fourier transformation of the autocorrelation cated on the plot.
function, or the direct method of splitting the
record into suitable blocks and subjecting these For tests in irregular waves, the correspond-
to Fast Fourier Transform. ing wave-energy spectrum should be defined.
(ω ) = 155 (ζ )
3. PARAMETERS ~ 2
− 944
exp 4 4 (3.3)
W 1/ 3 Y
SJ
3.1 Parameters to be Taken into Account T ω
1
4 5
T1 ω
(m²/s)
The amplitudes and phases of hull motions in where
six degrees of freedom 0.191ωT1 − 1 2
Wave height Y = exp −
Accelerations. 2σ
Relative motion
Rudder angle. and
Encounter angle σ = 0.07 for ω < 5.24 / T1
Leeway (or drift) angle σ = 0.09 for ω > 5.24 / T1
Impact pressures on the hull or on deck
Added resistance in waves This formulation can be used with other char-
Water on deck. acteristic periods by use of
Number of propeller revolutions, torque and
thrust. T1 = 0.924 T-1 = 0.834 T0 = 1.073 T2
Sea spectra
Where
T-1 is energy average period 2πm −1 / m 0
3.2 Recommendations of ITTC For Parame-
ters T0 is the peak period 2π / ω 0
where ω 0 is the circular frequency of the spec-
1975 Performance in irregular waves should trum peak.
be presented in non-dimensional form involv-
T1 is the average period 2πm 0 / m1
ing wave characteristics period and characteris-
tic wave height. T2 is the average zero crossing period,
2π m 0 / m 2
1978 Recommendation for open ocean spectral
formulation:
4. VALIDATION
A −B / ω 4
S( ω ) = e (m²/s)
ω5 4.1 Uncertainty Analysis
where
The detailed procedure of an uncertainty analy-
~ sis is shown in the appendix using the sample
A = 173(ζ W )1 / 3 / T14
2
anlysis of the S-175 ship.
B = 691 / T14 , T1 = 2πm0 / m1
7-6) Comparison of the Computer Calculations 14) Validation, Standards of Reporting and
of Ship Motions and Vertical Wave Bending Uncertainty Analysis Strip Theory Predic-
Moment (11th 1966 pp. 442-445) tions ( 19th 1990 pp.460-464)
Series 60, Cb=0.60 and 0.70
15) ITTC Database of Seakeeping Experi-
8) Comparison of the Computer Calculations ments (20th 1993 pp.449-451)
for Ship Motions and Seakeeping Qualities Two Dimensional Model, Wigley Hull
by Strip Theory (14th 1975 Vol.4pp.341- Form, S-175
350)
A Large-Sized Ore Carrier 16) Validation of Seakeeping Calculations
(21st 1996 pp.41-43)
9) Comparison on Results Obtained with Basic Theoretical Limitations
Computer Programs to Predict Ship Mo- Numerical Software Engineering Aspects
tions in Six Degrees of Freedom
(15th 1978 pp. 79-90) 17) ITTC Database of Seakeeping Experi-
S-175,Cb = 0. 572 ments (21st 1996 pp.43)
S-175, High Speed Marine Vehicle
10) Comparison of Results Obtained with
Compute Programs to Predict Ship Mo-
ITTC – Recommended 7.5 – 02
07 - 02.1
Procedures Page 9 of 16
Testing and Extrapolation Methods
Effective Date Revision
Loads and Responses, Sea Keeping 2002 01
Sea Keeping Experiments
The precision errors of wave amplitude, variables are obtained analytically using ship
heave and pitch amplitudes and wave encoun- motion program based on the strip method
ter frequency are obtained by (Park and Kang, 1989).
t
∂H ' ±
2 2
∂H ' ±
1 B ±
= Bζ a + BH (15)
∂ζ a
H'
∂θ ∂H
2 2 2
S ζ a = S + ∂θ S (12)
∂θ p θ p ∂θ θ t
t 2 2
∂H ' ∂H '
1 PH ' = Pζ a + PH (16)
∂ω e
2 2 ∂ζ a ∂H
S ω e = ST (13)
∂T
where S ζ p and S ζ t are the precision index for U H± ' = (B ) + (P )
± 2
H' H
2
(17)
the mean value of wave peaks and troughs re-
spectively. The similar notations are used for
The pitch RAO, bias limit, precision limit
heave and pitch motions. S T is the precision
and overall uncertainty are
index for the mean value of peak-to-peak peri-
ods. θ
θ'= (18)
360ζ a
A.1.2 Response Amplitude Operator (RAO) λ
and Overall Uncertainty 2 2 2
∂θ ' ' ∂θ ' ± ∂θ ' ±
B =
±
θ '' Bλ + Bθ + B ζ a
This section describes the procedure of ∂λa ∂θ ∂ζ a
combining all the errors obtained in A.1.1 and (19)
data reduction equations (DRE's) to get heave 2 2 2
and pitch RAO's and their overall uncertainties. ∂θ ' ∂θ ' ∂θ '
Pθ ''' = Pλ + Pθ + Pζ a
∂λ ∂θ ∂ζ a
For the variables such as LBP, B, d, KG, kyy
and model speed (Vm), the data reduction equa-
tions which describe the functional dependency U θ±' = (B ) + (P )
± 2
θ' θ'
2
(21)
of the heave and pitch RAO's to these variables
cannot be found as functional forms. Therefore,
the error propagation coefficients for these
ITTC – Recommended 7.5 – 02
07 - 02.1
Procedures Page 12 of 16
Testing and Extrapolation Methods
Effective Date Revision
Loads and Responses, Sea Keeping 2002 01
Sea Keeping Experiments
where the value of wavelength, λ , and its bias where ∂H ' / ∂ω ' and ∂θ ' / ∂ω ' are the slopes of
limit, B λ , and precision limit, Pλ can be ob- the heave and pitch RAO's respectively, as a
tained using the relation between the absolute function of ω .
frequency, ω , and the encounter frequency,
ω e , and wave dispersion relation in deep wa-
A.1.3 Results
ter.
Table 5 contains summary of the overall
For the comparison of RAO, nondimen- uncertainties of heave and pitch RAO's. The
sionalized absolute frequency, ω L / g , is ratios of overall uncertainties to heave RAO's
used as an independent variable. The value of range from ±2.0% to ±3.5% and for pitch
nondimensionalized absolute frequency, its RAO's they range from ±3.5% to ± 6.0%. At
bias and precision limits and overall uncer- higher encounter frequencies, as the magni-
tainty are tudes of the heave and pitch RAO's decrease,
the uncertainty limits tend to increase signifi-
2 2 cantly.
∂ω ' ∂ω '
Bω ' = BV + BL (22)
∂V ∂L Fig.1 and Fig.2 show the final combined
uncertainty limits of the heave and pitch
2 2 2 RAO's. Presented in these figures are the
∂ω ' ∂ω ' ∂ω '
Pω ' = PV + PL + Pω e analysis results of the RAO's using the strip
∂V ∂L ∂ω e theory indicated by the triangles (Park and
(23) Kang, 1989). The computed values of the
heave and pitch RAO's by the strip method
were found to lie far outside of the uncertainty
Uω' = (Bω ' )2 + (Pω ' )2 (24) limits of 95% confidence level. These differ-
ences between the test results and analysis re-
sults have been noticed in the report of the 15th
Combining Eqs. (17), (21) and (24), the fi- and 16th ITTC Seakeeping Committee Com-
nal overall uncertainties for heave and pitch parative Study on Ship Motion Program (1978,
RAO's become 1981).
2
∂H '
~
U H± ' = (U )
± 2
+
Uω' (25) A.1.4 Conclusions
∂ω '
H'
1) Uncertainty analysis has been successfully
∂θ '
2
applied to the motion test in regular waves.
~
U θ±' = (U )
± 2
θ' +
Uω' (26)
∂ω ' 2) 95% confidence intervals for heave and
pitch RAO's were approximately ± 3.0%
and ± 5.0% respectively.
ITTC – Recommended 7.5 – 02
07 - 02.1
Procedures Page 13 of 16
Testing and Extrapolation Methods
Effective Date Revision
Loads and Responses, Sea Keeping 2002 01
Sea Keeping Experiments