213-43556 LR Guidance Notes On Springing and Whipping v1 5 Draft.5 Draft

ShipRight
Design and Construction
Structural Design Assessment
Guidance Notes on the Assessment of Global

Design Loads of Large Container Ships and Other
Ships Prone to Whipping and Springing
June 2014
Draft Version
Guidance notes on the assessment of global design loads of large container ships and
other ships prone to whipping and springing
Contents
1 INTRODUCTION 4
1.1 Guidance Document 4
1.2 Application 4
1.2.1 General 4
1.2.2 Assessment of non-linear ship motion wave loads 4
1.2.3 Assessment of whipping 5
1.2.4 Assessment of springing responses including the effect on fatigue 5
1.2.5 Symbols 5
1.3 Non-linear effects 6
1.4 Whipping Phenomena 6
1.5 Springing 7
1.6 Hull girder natural vibration frequencies 8
1.7 Assessment Procedure 8
1.7.1 Assessment of the long term design linear wave loads 8
1.7.2 Assessment of the non-linear ship motion wave loads 8
1.7.3 Assessment of the hull girder loads including whipping loads 9
1.7.4 Assessment of fatigue including hull girder springing loads 9
1.8 Documentation and plans to be submitted 9
2 ASSESSMENT OF THE LONG TERM LINEAR DESIGN WAVE LOADS 11
2.1 General 11
2.2 Loading conditions 11
2.2.1 Loading conditions for the whipping and non-linear assessments 11
2.2.2 Loading conditions for the springing assessment 11
2.3 Hydrodynamic model 12
2.4 Verification of hydrodynamic models 12
2.5 Roll damping coefficient 13
2.6 Ship operational conditions 13
2.7 Ship motion and load responses in frequency domain 14
2.8 Wave environment 14
2.9 Linear design wave bending moment 15
2.10 Long term calculation 15
3 ASSESSMENT OF NON-LINEAR SHIP MOTION WAVE LOADS 16
3.1 General 16
3.2 Determination of the Equivalent Design Wave (EDW) 16
3.3 Determination of the Equivalent Design Sea State (EDS) 17
3.4 Computation of Vertical Wave Bending Moments in Time Domain 18
3.4.1 Wave Components of the EDS approach 18
3.5 Derivation of hogging and sagging non-linear factors for the vertical wave bending moment 19
3.5.1 Time domain simulation using the EDW approach 19
3.5.2 Time domain simulation using the EDS approach 20
3.6 Design vertical wave bending moments 20
3.7 Application to the structural assessment 20
4 ASSESSMENT OF THE HULL GIRDER LOADS INCLUDING WHIPPING LOADS 22
4.1 General 22
4.2 Determine EDW and additional equivalent design waves 22
Lloyds Register 1 June 2014

MTES/SAH Draft v1.5
4.3 Derivation of the EDSs 23

4.4 Loading condition 23
4.6 Structural model of the ship 23
4.6.1 2D structural modelling 23
4.6.2 3D structural modelling 24
4.6.3 Eigenvector analysis techniques 24
4.6.4 Loading a whole ship 3D FE model 25
4.6.5 Structural damping 26
4.7 Hydrodynamic definition of the bow flare and slamming regions 26
4.7.1 Impact load evaluation 26
4.7.2 Definition of the bow flare impact sections 27
4.8 Calculation of whipping vertical bending moments in time domain 27
4.9 Derivation of hogging and sagging whipping factor for the vertical wave bending moment 28
4.9.1 Time domain simulation using the EDW approach 28
4.9.2 Time domain simulation using the EDS approach 29
4.10 Design vertical wave bending moments 31
4.11 Application to the structural assessment 32
4.11.1 Ultimate strength assessment 32
5 STATISTICAL POST PROCESSING 34
5.1 Evaluation of the probable maximum value in 3 hours 34
6 ASSESSMENT OF FATIGUE INCLUDING HULL GIRDER SPRINGING LOADS 36
6.1 General 36
6.2 Outline of Assessment Procedure 36
6.2.1 Assessment of linear stress RAOs for all headings and speeds including springing responses 37
6.2.2 Assessment of linear bending moment RAOs for all headings and speeds including springing responses 38
6.2.3 Assessment of non-linear springing responses for critical wave frequencies 39
6.3 Springing Fatigue Reduction (SFR) factor 39
6.3.1 Overview 39
6.3.2 Calculation of fatigue damage index including and excluding springing for one sea state 40
6.3.3 Fine mesh details and Stress Concentration Factors (SCF) 41
6.3.4 Calculation of the final fatigue life and damage index 41
6.4 Calculation of springing responses 43
6.4.1 Hydroelastic methods 43
6.4.2 Springing responses using linear frequency domain analysis 44
6.5 Loading condition 44
6.7 Structural model of the ship 44
6.8 Structural damping 44
6.9 Calibration of springing responses 44
7 ALTERNATIVE NON-LINEAR SPRINGING FATIGUE ASSESSMENT METHODS 47
7.1 Springing responses using time domain analysis 47
7.2 Assessment of the springing response to critical wave frequencies 48
7.2.1 Overview of springing fatigue method using the critical wave frequency method 48
7.2.2 Critical wave frequencies for excitation of hull girder vibration modes 50
7.2.3 Example of selection of critical wave frequencies 52
7.2.4 Calculation of the total rigid body and springing response for the fatigue analysis. 53

MTES/SAH Draft v1.5
List of Figures
Figure 1 Typical non-linear effects of VBM due to hull shape in a model test. 6
Figure 2 Typical whipping effects of VBM in the full scale measurement. 7
Figure 3 Flowchart of the main elements of the non-linear design load assessment procedure 9
Figure 5 Vertical bending moment amidships, = 0.91Lpp 16
Figure 6 Typical time series of VBM amidships 22
Figure 7 Example of modelling containers figure will be improved. Ideally each 40ft Bay should be modelled 26
Figure 8 Example of cross-sections defined for bow flare slamming 27
Figure 9 Example of convergence of whipping hog factor with duration of simulation time 28
Figure 10 Counting method for extremes of the linear, non-linear and whipping response time domain traces 30
Figure 11 Weibull fitting of the linear, non-linear and whipping vertical sagging bending moment distributions 31
Figure 12 Weibull fitting of the linear, non-linear and whipping vertical hogging bending moment distributions 31
Figure 13 Example of good and poor fitting of Weibull distributions 34
Figure 14 Measured and calculated vertical bending moment amidships for a JONSWAP wave spectrum 36
Figure 15 Stress RAO including springing effects for a very flexible ship 38
Figure 16 Typical stress RAO curve and cut-off frequency, Head sea case showing the dominant vertical bending
mode 41
Figure 17 Typical stress RAO curve and cut-off frequency, Oblique sea case showing the dominant vertical
bending mode and small torsional or horizontal vibrational modes 41
Figure 18 Example of the effect of heading, wave height and period and ship speed on the contribution of
springing effects on fatigue. 43
Figure 19 Flowchart of the assessment of fatigue using RAOs including springing responses 46
Figure 20 Calculations of time domain springing responses for a regular wave 48
Figure 21 Flowchart of the springing response to the critical wave frequencies and the application to the fatigue
assessment 50
Figure 22 Example of the total response energy spectrum including springing (figure to be improved) 54
List of Tables
Table 1 Recommended size of hydrodynamic panels for whipping and springing ship motion analyses 12
Table 2 Ship speed versus wave height relationship for container ships 13
Table 3 Ship motion analysis parameters 14
Table 4 Probability of sea-states in the North Atlantic described as occurrence per 100000 observations. 14
Table 5 Example of individual probability contribution of each sea states to the long term 10-8 value. 18
Table 6 Example of the critical wave frequencies for exciting different vibration modes. 52
Table 7 Example of the selection of critical wave frequencies. 52

MTES/SAH Draft v1.5
1 Introduction
1.1 Guidance Document
This Guidance Document provides guidance and recommendations on the assessment of non-linear effects and hull
girder flexibility, so whipping and springing actions, on the vertical wave bending moment and other global loads.
These non-linear effects, especially the whipping responses, are complex issues as there are many aspects that are
difficult to address explicitly by calculation or testing. This document provides guidance on appropriate methods for the
assessment of the non-linear effects induced by hull shape, wave impact loads and hull girder flexibility issues. The
proposed methods that the designer intends to adopt are to be discussed and agreed with Lloyds Register prior to
commencement of the assessment. Detailed reports of the assessment are to be submitted to Lloyds Register for design
review.
The results of the non-linear wave induced ship motion loads analysis, the springing or whipping assessment may need
to be applied to the Structural Design Assessment (SDA) and Fatigue Design Assessment (FDA) procedures. The
application will need to be specially considered for ships that are sensitive to springing, have large variation in hull
shape above and below the waterline, or subject to significant whipping loads. For these cases it may be necessary to
review the structural integrity using the Response Based Analysis (RBA) or similar, that is direct application of
dynamic loads from a ship motion analysis to the Finite Element Analysis. The method of application of dynamic loads
to FEA is to be discussed and agreed with Lloyds Register prior to commencement.
1.2 Application
1.2.1 General
This document is particularly applicable to ultra large container ships, Great Lakers, ultra large ore carriers, ultra large
bulk carriers as well as other ships which are likely to be prone to whipping or springing due to either high L/D ratios,
high speed or high values of bow or stern flare regions.
These guidance notes are applicable to assess the vertical wave bending moment (and global loads induced by waves) to
cover non-linear effects of hull form shape including whipping due to the slamming and operating worldwide in any sea
condition. The notes include details on how to assess the hull girder strength against whipping induced hull girder loads.
These guidance notes are applicable to assess the excitation of natural modes of vibration of the hull girder (springing)
by waves and for the analysis of springing loads for the fatigue assessment.
The document is also applicable to other hull girder loads such as the wave torsional moment and horizontal wave
bending moment, these issues are not explicitly covered in this document but the proposed methods are equally
applicable.
The assessment of non-linear ship motion wave loads, whipping effects or effects of springing are to be in accordance
with the rule requirements. In addition, any of these may be required either on a voluntary basis or when stipulated by
Lloyd's Register when it considers it necessary.
The following paragraphs give examples of when these issues may need to be investigated.
1.2.2 Assessment of non-linear ship motion wave loads
An analysis of the non-linear ship motion wave loads of a ship is likely to be required when one or more of the
following criteria are applicable:
a) |ffS| > 1,4 or

MTES/SAH Draft v1.5
b) RABF > 0,2 or

c) RABFU > 0,2
d) Cb < 0.6
1.2.3 Assessment of whipping
An analysis of the whipping responses of a ship is likely to be required when one or more of the following criteria are
applicable:
a) fc > fsp
b) L > 400 m
c) L > 300 m and one or more of
i) |ffS| > 1,4 or
ii) RABF > 0,2 or
iii) RABFU > 0,2
d) L/D > 15
1.2.4 Assessment of springing responses including the effect on fatigue
An analysis of the hull girder springing responses is likely to be required when one or more of the following conditions
applies:
a) L > 250 m and fc > fsp
b) Use of HT47 or above for the deck or hatch side coaming
c) Use of HT36 or above for the bottom shell
If the springing assessment shows that there is a significant springing response, then it may be necessary to undertake a
more detailed fatigue analysis including springing effects. The fatigue assessment is to be carried out in accordance
with the appropriate ShipRight FDA procedure with the results adjusted to take account of springing.
1.2.5 Symbols
ffS is defined in Pt 4, Ch 2, 2.4.1
RABF is the bow flare area ratio for the lower region just above the still waterline
ABF

0,1LB
= WL
RABFU is the bow flare area ratio for the upper region near the deck
ABFU

0,1LB
= WL
ABF ,BWL are defined in Pt 4, Ch 2, 2.4.1
ABFU is the bow flare area in m2 for the region from a waterline of TC,U to TC,2U, calculated in the same way
as ABF see Pt 4, Ch 2, 2.4.1
TC,2U = T + C1 m or the local deck edge height if this is lower
C1 is defined in Table 4.5.1 in Pt 3, Ch 4
fc is a wave encounter frequency at which it is expected that whipping or springing will become
important
1,4 1,4V
1
2 2g Hz
=
fsp is the natural frequency of the 2 node hull girder vertical bending mode in Hz. For ships of normal
dimensions, this can be very approximately calculated as:

MTES/SAH Draft v1.5
1,1 E I10 5
=
2
Hz
L f am B Td C b
V = required operational speed in knots, Usually this will be as defined in Pt 3, Ch 1,6.1.10
E = Young's modulus in N/mm2
= 206000 N/mm2 for steel
I is the midship moment of inertia in m4, see Pt 3, Ch 4
Td is the required operating draught, in metres.
fam = 1,8 for container ships. For other ship types, then this value will be specially considered. Typically it
will be in the range 1,8 to 2,1.
Cb, L,B and D are given in Pt 3, Ch 1,6.
1.3 Non-linear effects
For large ships and ships of unusual hull form proportions, then the standard rule values of global hull girder loads are
not 100% reliable and direct calculation of these responses is required. As part of this assessment, the effects of high
forward speed and the variation of the above and below water plane hull shape need to be addressed. Hence a non-linear
ship motion analysis is necessary to determine the appropriate design loads. This will provide revised hogging and
sagging correction factors for application to the linear wave bending moments and shear forces. It also allows of the
effects of hull shape, non-linear wetted area to be taken into account in the calculation of the horizontal bending and
torsional moments.
A model test time history of vertical wave bending moment of a 6500 TEU containership in a regular wave with a 10 m
wave height is shown in Figure 1. This illustrates the typical non-linear effects due to the hull shape.
4.0E+06
3.0E+06
2.0E+06
VBM (kNm)
1.0E+06
0.0E+00
-1.0E+06
-2.0E+06
-3.0E+06
-4.0E+06
300 320 340 360 380 400
Time (sec)
Figure 1 Typical non-linear effects of VBM due to hull shape in a model test.
1.4 Whipping Phenomena
Whipping of a ship is the rapid flexing of the hull girder as a consequence of wave impacts on the hull. These high
frequency oscillations of the hull girder result in increased vertical wave bending moments and shear forces. The
dominant oscillation mode is at the period of the first (2-node) natural vertical vibration mode of the hull girder. Other
hull girder vibrational modes may also be excited, such as the first (1-node) torsional mode or the coupled 1-node
torsional and 2-node horizontal bending mode.

MTES/SAH Draft v1.5
High whipping responses are usually driven by bow flare impacts due to large bow flare angles and high speed or by
bottom slamming. Occasionally stern counter slamming can lead to high whipping responses. The oscillations of the
whipping responses usually decay rapidly after several wave periods due to damping effects. Ships which have hull
girder natural frequencies close to the wave energy region are potentially prone to show prolonged springing after an
impact event as there is little damping resistance close to the dominant hull girder vibration mode and the continuous
wave excitation keeps springing going.
The increase in vertical bending due to whipping is primarily a global strength issue. It is not a fatigue issue as the
whipping induced vertical bending moment oscillations usually damp out quickly and hence the number of cycles is
small.
A full scale measurement time history of vertical wave bending moment of an 8500 TEU containership with a typical
whipping response in waves is shown in Figure 2.
6.0E+06
4.0E+06
2.0E+06
VBM (kNm)
0.0E+00
-2.0E+06
-4.0E+06
-6.0E+06
220 240 260 280 300 320 340
Time (sec)
Figure 2 Typical whipping effects of VBM in the full scale measurement.
1.5 Springing
Springing of a ship is the continual vibration of the hull girder as a consequence of the waves exciting resonant hull
girder frequencies. This flexing of the hull girder due to springing may continue for several hours once initiated.
Springing is an issue for ships which have low natural vibration frequencies of bending or torsional modes, typically
when the lowest natural frequency is less than 3 rad/sec ( 0.5 Hz) and the ship speed is above 20 knots. This is the case
for large container ships due to their high speed and open cross sections. Great Lakers are very prone to springing due
to their low second moment of inertia and very high L/D ratio, which result in low natural frequencies of vertical
vibration modes. Full scale measurements have shown that large bulk carriers are also sensitive to springing.
The magnitude of the springing bending moment and hence bending stresses is usually low, however the number of
cycles is very large (about 4-8 times the number of wave cycles) and this can make springing important for fatigue.
Springing is not normally a strength issue, but it can have an important effect on the fatigue life of a structure. Studies
have shown that the fatigue life for the most sensitive details may be reduced by a factor of 30% or more for container
ships of around 6000TEU and this will be more critical for larger ships, hence it is necessary to review the effects of
springing for large container ships to ensure adequate fatigue life.

MTES/SAH Draft v1.5
1.6 Hull girder natural vibration frequencies
All structural systems have numerous natural vibration frequencies and any cyclic load and its harmonics that are near
to these natural frequencies will result in these natural vibration modes being excited. In addition, these natural
vibration modes may be excited by impulsive loads such as bottom slamming and bow flare impacts which have a wide
excitation range in the frequency spectrum.
For a ship with a full closed deck structure, e.g., tanker, the two-node vertical bending vibration mode, which has the
lowest natural frequency, is the most important mode to be considered in the springing/whipping analysis. This
vibration mode has a small damping ratio and is closest to the sensitive frequencies of waves in the real sea state hence
this vibration mode is the one most often observed in service.
For ships with open deck structures, such as container ships and bulk carriers, then the coupled 1-node torsional
vibration mode and the 2-node horizontal bending mode may have a lower natural frequency than the 2-node vertical
bending mode. Hence this coupled vibration mode can be important.
Other possible important vibration modes for these ships are the higher order (3-node) vertical bending and combined
horizontal and torsional modes. These higher order modes have higher natural frequencies and high damping effects and
are therefore rarely observed in full scale measurement.
The natural frequencies of ship hull girder are governed by the ship length (L), hull girder stiffness (EI) and mass/added
mass. For vertical bending modes, the natural frequencies are typically given by the following relationship:
1 EI
f Hz
L2 M
Hence high ship length (L) leads to lower natural vibration frequencies and increases susceptibility to springing
generally.
1.7 Assessment Procedure
The process to be followed for the assessment of non-linear design wave bending moment and other global loads due to
non-linear effects and whipping response is outlined in Figure 3.
1.7.1 Assessment of the long term design linear wave loads
Section 2 provides the conventional method to determine the linear hull girder design loads along the ship length for
conventional ships. This involves calculating the ship motions and global loads for a range of ship loading conditions,
speeds, heading angles and wave periods. The ship motions are calculated using a linear potential flow hydrodynamic
program and statistical (short term and long term) results using North Atlantic wave scatter diagram.
1.7.2 Assessment of the non-linear ship motion wave loads
Section 3 provides a method to determine the non-linear hull girder design loads due to hull shape issues. The method
provides revised non-linear sag and hog factors which can be applied in the rule assessment process.
A consistent method is provided that is based on calculating suitable Equivalent Design Sea States (EDS) appropriate
for severe sea states and undertaking non-linear time domain simulations for the ship motions and vertical wave
bending moments. The method will also allow calculation of the effects of nonlinearities on vertical shear forces,
horizontal bending and torsional moments. An Equivalent Design Wave (EDW) is also possible, but it is recommended
that the EDS approach is adopted.

MTES/SAH Draft v1.5
Linear Design Loads

Loading
Speed Heading
condition
Ship Motion and

Load Analysis in
Regular Waves
Environmental Linear Design

analysis Loads
Nonlinear Design Loads Nonlinear Design Loads

due to Hull form Shape due to Whipping Resonse
Modal Analysis
of Ship Hull
Structure
Ship Motion and Ship Motion and
Load Analysis in EDW/EDS EDW/EDS Load Analysis in
time domain time domain
Slamming
Force
Calculation
Nonlinear Sag/ Whipping Sag/

Hog Ratio hog factors
Figure 3 Flowchart of the main elements of the non-linear design load assessment procedure
1.7.3 Assessment of the hull girder loads including whipping loads
Section 4 provides a method to determine hull girder design loads taking account of non-linear issues, wave impact
loads and hull girder flexibility. The method provides a whipping enhancement factor for the rule vertical wave bending
moment which can be applied in the rule assessment.
A consistent method is provided that is based on calculating suitable EDW/EDS appropriate for severe sea states and
calculation of the hull girder natural vibration modes. This is followed by undertaking non-linear time domain
simulations using a hydroelastic ship motion code which can also calculate wave impact loads and hence the resulting
vertical wave bending moments.
1.7.4 Assessment of fatigue including hull girder springing loads
Section 6 provides a method to evaluate the springing response of a ships hull girder for any sea state. The resulting
response can then be used in a fatigue analysis to predict the fatigue life including springing.
Due to complexity of this issue, several methods are proposed and the designer can select the most appropriate for their
own internal fatigue assessment procedures. The proposed methods are based on assessment of springing in sea states
appropriate for moderate wave environments where springing responses are expected to be significant and sea states
which are expected to be dominant for the predicted fatigue life.
Consistent methods are presented which use either frequency domain or time domain simulations based on a
hydroelastic ship motion code to predict the springing responses. As the magnitudes of the springing responses are
difficult to predict, then the assessment of springing needs to be undertaken on the basis of a comparative assessment
where the springing response of the new design is compared with the springing response of a known benchmark design.
1.8 Documentation and plans to be submitted

MTES/SAH Draft v1.5
Reports are to be submitted to Lloyds Register for the approval of the hull girder structure with regard to whipping and
springing and are to contain:
list of plans used, including dates and versions;
detailed description of structural modelling, including all modelling assumptions;
plots to demonstrate correct structural modelling and assigned properties;
details of material properties used;
details of selected loading conditions;
details and plots of the hydrodynamic model including mesh density, mass buoyancy balance parameters,
calculated SWBM distribution;
plots and tables of eigenvector results that demonstrate the correct behaviour of the structural model;
details of the dry and wet natural frequencies for the critical vibration modes, demonstration that all the critical
vibration modes have been captured;
details of the linear ship motion analysis and the selection of the equivalent design sea states;
selected time domain traces of the linear, non-linear and whipping analyses;
statistical analysis process and results;
ultimate strength assessment and results;
fatigue assessment procedure, assumptions and results;
springing stress RAOs or BM plots to demonstrate the springing effects have been captured
demonstration that the structure complies, or otherwise, with the design criteria; and
proposed amendments to structure where necessary, including revised assessment of results.

MTES/SAH Draft v1.5
2 Assessment of the long term linear design wave loads
2.1 General
In this chapter, detailed requirements are given to derive the long term design linear wave loads using a first principles
direct calculation procedure.
2.2 Loading conditions
2.2.1 Loading conditions for the whipping and non-linear assessments
For a conventional container ship which always has a large hogging still water bending moment (SWBM), then the
loading condition to be used for the strength assessment is a full loading condition at the maximum normal operational
draft (usually the design draught) that maximises the hogging SWBM amidships. If it is expected that the ship will
operate at a deeper draught, for example the scantling draught, on a regular basis, then this deep draught condition
should also be assessed.
In some cases it may be necessary to also review the normal ballast loading condition or a typical container loading
condition at a lighter draught that maximises the sagging SWBM amidships (or minimises the hogging SWBM).
Notes
It is common for container ships to have a "scantling draught" and a "design draught" specified. The
"scantling draught" often being used to get additional scantlings due to the way that draught is used in the
local scantling rule equations. The choice of draught has little effect on the rule design wave bending moments,
but a higher draught will usually result in an increase in the actual hogging SWBM.
However whipping bending moments can be significantly affected by the selection of draught as it changes the
forward draught. Increasing the draught forward affects the bow slamming pressures as it might bring the
higher bow flare region closer to the waterline. In some cases this has been found to increase the whipping
bending moment.
Consequently it is necessary to clearly define the correct design draught for the container ship and use this
consistently throughout the design.
For other ship types, then a typical operating loading condition that maximises the hogging SWBM and a condition that
maximises the sagging SWBM are to be reviewed in association with draughts that are likely to result in maximum bow
flare impact pressures.
It is not necessary for the permissible SWBM to be matched in any of the above loading conditions if it is not
realistically possible to achieve the permissible value. However the results of the whipping assessment may require a
reduction of the rule permissible SWBM in order to ensure that the ship complies with the whipping assessment.
Note
In practice the worst SWBM at the specified draught is to be used, hence the ballast condition SWBM values
should not be applied to a design draught loading condition. It also means that the Sagging SWBM will be
assigned as a large hogging value as it is container ships always hog in a fully loaded condition.
2.2.2 Loading conditions for the springing assessment
The loading condition to be used for the springing and fatigue assessment is a typical full loading condition at the
design draught.

MTES/SAH Draft v1.5
In some cases it may be necessary to also review a normal ballast loading condition or a typical container loading
condition at a lighter draught depending on the expected operational profile of the ship.
For other ship types, then a typical full load condition and a lightly loaded or ballast condition are to be reviewed. The
loading conditions to be chosen are to be representative of those most likely to be used during everyday operation.
2.3 Hydrodynamic model
Wave load and ships motion responses to unit regular waves are to be calculated using a linear potential flow
hydrodynamic program.
The mesh size should be based on the highest wave encounter frequency considered. Ideally, the length of the panels
should be such that one obtains at least 5 panels per wave length for the highest wave encounter frequency ( e max ),
i.e.:
2g
L panel
5e2max
Table 1 gives the minimum recommended panel size for various ship speeds and maximum frequencies to be assessed.
This illustrates the severe problem with satisfying the optimum panel size criteria. Hence consideration will be given to
relaxing this requirement for very large hull forms at high speed, but sensitivity studies on panel size will be necessary
to confirm that the results are converged.
Table 1 Recommended size of hydrodynamic panels for whipping and springing ship motion analyses
Maximum wave frequency to be assessed in rad/s
speed knots (V) 1.4 rad/s 1.8 rad/s 2.2 rad/s and above
5 3.4 m 1.8 m 1.0 m
15 1.4 m 0.7 m (1) 0.3 m (1)
20 1.0 m 0.5 m (1) 0.2 m (1)
(1) (1)
25 0.8 m 0.3 m 0.2 m (1)
Notes
1 Due to current computer hardware limits, the optimum size for the hydrodynamic panels will not be
achievable for large container ships. Hence a panel size of 1 m is recommended if possible.
2 For container ships, the maximum wave frequency to be assessed should be greater than the wave
frequency that corresponds to encounter wave frequency e =n, where n is the 2 node vertical bending
vibration frequency. The maximum wave frequency is given by:
2
2g 2g 2g
4 e
V V V

2
For other ship types, the maximum frequency may also need to consider other vibrational modes.
It is not normally necessary to consider a maximum wave frequency greater than 2.5 rad/s due to the very
low wave energy content above this frequency.
2.4 Verification of hydrodynamic models
Hydrostatic calculations based on the hydrodynamic mesh should be performed for each required loading conditions to
validate the hydrodynamic model(s).

MTES/SAH Draft v1.5
The following hydrostatic parameters are to be calculated and verified against the values given in the ship's loading
manual or calculation obtained from the ship's loading computer software. The following tolerances are to be met:
Displacement 0.5 % but ideally this should be less than 0.15 %
LCB - LCG 0.1m
Draught (Forward; Aft) 0.02m
SWBM 2%
Note.
It is advisable to ensure the difference between the summations of the mass distribution and the hydrodynamic
buoyancy are negligible. If the ship motion software does adjust the mass or buoyancy to achieve a perfect
balance, then the user should understand this could result in a large error in the hydrodynamic results.
The methods of ensuring a balance between the hydrodynamic mesh, in preferred order, are as follows:
1. adjust the hull form or draft and trim to achieve the mass and LCG
2. adjust the mass distribution to achieve the hydrodynamic displacement and LCB
3. scale the mass and move all mass items to achieve the hydrodynamic displacement and LCB
2.5 Roll damping coefficient
Viscous roll damping has a great influence on the ships roll motion at near roll resonance frequency. The viscous
damping effect is to be considered in the hydrodynamic analysis as follows:
The non-linear roll damping effect is to be considered in critical sea conditions where the encountered wave
frequencies are close to the roll resonance frequency.
The effects of bilge keel and other appendages are to be considered.
The effects of active roll stabilisers or roll damping tanks are to be ignored.
The values of roll damping factors are to be suitable for the analysis being undertaken. For the whipping analysis, then
the roll damping is to be applicable for severe sea states. For fatigue, then the roll damping is to be applicable for
moderate sea states. Suitable design sea states for the evaluation of the roll damping can be taken as the Equivalent
Design Sea state specified in the whipping and springing chapters.
2.6 Ship operational conditions
The ship speed is to be adjusted according to the significant wave height. For container ships the speed versus wave
height relationship is to be assumed as given in Table 2.
Table 2 Ship speed versus wave height relationship for container ships
Significant Wave Height (Hs) Ship Speed
Hs > 10.5m 25% Vs
10.5m > Hs > 7.5m 50% Vs
7.5m > Hs > 4.5m 75% Vs
Hs < 4.5m Vs
Vs Maximum service speed
A different relationship is to be assumed if the ship is of unusual form or installed power or otherwise specified by the
owner/designer
The speed/wave height relationship table is to be applied for the selection of the appropriate ship speed for the whipping
and springing assessments. If the equivalent design wave approach is being adopted, then 5 knots is to be used for the
whipping analysis and service speed is to be used for the springing analysis.

MTES/SAH Draft v1.5
2.7 Ship motion and load responses in frequency domain
Ship's motion and load responses to sinusoidal waves are to be calculated to determine the Response Amplitude
Operators (hereinafter referred as RAOs) and the associated phase angles for each of the ship's loading conditions
specified.
The RAOs are to be calculated for the following parameters specified in Table 3
Table 3 Ship motion analysis parameters

Comment
Loading conditions As specified in 2.2
Speed As specified in 2.6
Wave/ship heading Head seas (= 1800) to following seas (= 00) depending on the
source of wave environment data. In general, at 15 degree
increments.
Wave frequency Upper limit of frequency range depends on the maximum ships
speed and ships length and wave environment.
In general, 0.2 to 1.5 rad/s at a maximum increment of 0.05 rad/s
For each ship loading condition, the following wave load responses are to be calculated:
VWBM, VWSF, HWBM, HWSF from 0.1 to 0.9L in 0.1L increments
Torsion moment about shear centre from 0.1 to 0.9L in 0.1L increments
The position of the Shear centre will need to be specially considered. The geometric 2D shear centre is
only an approximation and a more representative shear centre that takes account of 3D effects should be
considered. This could be derived after assessment of the torsional responses using a whole ship 3D FE
or LRPASS 20204 or similar.
2.8 Wave environment
The standard wave environment to be used for the assessment of the long term loads acting on ships which are
operating worldwide is the North Atlantic all-directions scatter diagram as specified in IACS Rec. No. 34, see Table 4.
This wave environment specifies the probability of occurrence of individual sea states, each sea state being defined by a
significant wave height (HS) and a zero up-crossing period (TZ). Each sea state is to be characterised using the modified
Pierson-Moskowitz wave energy spectrum and is to be assumed to exist for a duration of 3 hours.
Table 4 Probability of sea-states in the North Atlantic described as occurrence per 100000 observations.

MTES/SAH Draft v1.5
The modified Pierson-Moskowitz wave energy spectrum, also called the ISSC spectrum, is to be used to model the
wave environment in the North Atlantic or any open ocean area. The assumption of short crested seas with a cosine
squared spreading function should be applied.
The IACS Rec. no. 34 wave environment is to be used for the whipping assessment.
The wave environment for a restricted service requirement will be subject to special consideration.
The wave environment for springing fatigue assessment is to be derived in accordance with the FDA2 procedure
including adoption of the appropriate ship speed to wave height relationship.
2.9 Linear design wave bending moment
The linear design vertical wave bending moment, VBMLT, is to be calculated by the long term statistical method using
the wave scatter data described in 2.8 with a return period of at least 20 years, corresponding to about 10-8 probability of
exceedence.
The standard Rule value of the linear wave bending moment Mwo is given by Part 3, Ch 4, 5.2.1 of Lloyd's Register's
Rules and Regulations for the Classification of Ships.
2.10 Long term calculation
The long term approach is an accumulation of the statistics for all short term sea conditions taking into account the
frequency of occurrence of each short term sea state. When a ship travels at constant speed, the long term probability of
~
exceedence Q ( x ) for x is
360
1 Q ( x H , Tz ) f ( H s , Tz ) dH s dT z d
~
Q( x ) 1 s
0 0 0
Where
f ( H s , Tz ) is the joint probability density function of significant wave height, Hs, and average zero-crossing
wave period, Tz, as given by the wave scatter diagram of IACS Rec. 34, see Table 4. Note the
IACS standard assumption is to assume equal probability of heading.
Q( x H s , Tz ) is the probability of exceedence for x in a single short term sea state with significant wave
height and averaged zero-crossing wave period ( H s , Tz ) .

heading angle
It is generally assumed that the short term response is a narrow band random process (with bandwidth parameter 0 )
and its probability of exceedence follows the Rayleigh distribution, that is
2

1
Q( x H s , Tz ) exp
m
,
2 0

and m0 is the response variance of the single sea state from short term analysis.
Alternatively the long term values may be derived considering suitable bandwidth correction factors.

MTES/SAH Draft v1.5
3 Assessment of non-linear ship motion wave loads
3.1 General
In this section, requirements are given to derive the design wave bending moment to cover non-linear effects induced by
hull shape using first principles direct calculation procedure. It is also necessary to derive the longitudinal distribution
of the wave bending moment in order to ensure that the hull girder strength of the ship is valid along its length.
The process to derive the non-linear response of vertical shear force, horizontal bending moment and shear force and
torsional moment is the same as that to derive the non-linear vertical bending moment response.
The procedure derives the design wave bending moment based on either the Equivalent Design Sea states (EDS) or
Equivalent Design Wave (EDW) concept. The recommended concept is the EDS method. The EDW method is useful
for initial checking and review, but is not recommended as it is harder to achieve consistent results.
Note
It is only necessary to apply one of these two methods and LR's preferred method is to apply the EDS concept.
At the moment, whipping and springing analysis methods are under development and all organisations
involved in this topic are still refining procedures, hence these guidance notes present the best methods that
are currently considered viable.
The design values are then given by applying this EDS (or EDW) to a non-linear time domain ship motion analysis. It
should be noted that different longitudinal locations and different hull girder responses may well require different EDSs
or EDWs.
The procedure to derive the non-linear response for other load values such as horizontal bending moments and torsional
moments is similar.
Figure 4 shows a time domain linear wave bending moment response to a regular wave of length 0.91L, curve labelled
linear. The nonlinear curve shows the non-linear ship motion response due to hull shape and actual wave profile.
This shows that the hogging bending moment (positive values) are largely unchanged whereas the sagging bending
moments are significantly increased. The whip curve shows the combined effect of non-linear responses and
whipping on the vertical bending moment.
Linear nonlinear whip
1.5E+07
1.0E+07
5.0E+06
VBM, kNm
0.0E+00
-5.0E+06
-1.0E+07
-1.5E+07
-2.0E+07
60 70 80 90 100 110 120
time, sec
Figure 4 Vertical bending moment amidships, = 0.91Lpp
3.2 Determination of the Equivalent Design Wave (EDW)

MTES/SAH Draft v1.5
Please note the recommended method is to use the EDS approach.
The EDW is a regular wave that induces the same design linear vertical bending moment as specified by the rules, Mwo
(see Pt 3, Ch 4.5 or Pt 4, Ch 2.4). As an alternative, it is acceptable to derive the EDW based on the Long Term (LT)
value, VBMLT, calculated in accordance with Section 2, in this case the long term value is to be used instead of the rule
value in to define MVL.
The period of the EDW is selected as the wave period of the peak of the RAO curve for vertical bending moment
amidships in head seas. The design wave height is determined so that it reproduces the rule (or LT) vertical bending
moment amidships.
EDW period = Peak period of VBM RAO curve in head seas
EDW Amplitude = Rule (or LT) design VBM / Max. amplitude of VBM RAO in head seas.
The calculation of the non-linear responses is then undertaken using a non-linear time domain ship motion analysis
using a regular sine wave.
However in order to review the effects of speed, the following additional equivalent design waves (hereinafter referred
as AEDWs) should also be reviewed:
AEDW with the same EDW amplitude, wave period = (EDW period one) seconds, 5 knots
AEDW with the same EDW amplitude, wave period = 2Lc g seconds, 5 knots
AEDW with the same EDW amplitude, wave period = 2Lc g 1 seconds, 5 knots
AEDW with 65% of EDW amplitude, wave period = 2Lc g seconds, 50% service speed
AEDW with the 65% of EDW amplitude, wave period = 2Lc g 1 seconds, 50% service speed
AEDW with 35% of EDW amplitude, wave period = 2Lc g seconds, service speed
AEDW with 35% of EDW amplitude, wave period = 2Lc g 1 seconds, service speed
where
Lpp ship length between perpendiculars, in m.
Lc approximate critical wavelength, in m, to be taken as 0.9Lpp
It is necessary to review the sensitivity of response to small variations in encounter wave period, so extra AEDWs may
need to be selected. The designer is to demonstrate that the selection of speed, wave amplitude and wave period is such
that the maximum non-linear response has been determined.
3.3 Determination of the Equivalent Design Sea State (EDS)
The EDS is defined as the sea state which has the maximum contribution to the 10-8 probability of exceedence long
term vertical wave bending moment amidships. The LT value is calculated in accordance with Section 2.
The basic steps to determine the EDS are to be follow:

(1) carry out the long term statistical analysis for the vertical wave bending moment amidships, determine the
~
specific vertical wave bending moment amplitude m at a probability of exceedence Q( x m ) equal to 10-8
(2) calculate the probability of exceedence Q( x m H s , Tz ) f ( H s , Tz , ) for every single sea state of the
wave scatter diagram based on the specific VBM value m . For vertical wave bending moments, then it is
adequate to only consider the head sea heading.
(3) the EDS is the sea state with maximum probability value, see example in Table 5

MTES/SAH Draft v1.5
The calculation of the non-linear responses is then undertaken using a non-linear rigid body time domain ship motion
analysis with this irregular EDS.
Table 5 Example of individual probability contribution of each sea states to the long term 10-8 value.
12.5 2%
11.5 4% 1%
10.5 1% 6% 2%
Hs
9.5 3% 4% 3%
8.5 2% 3% 3%
7.5
7.5 8.5 9.5 10.5 11.5
Tz
3.4 Computation of Vertical Wave Bending Moments in Time Domain
The predicted vertical wave bending moments of the EDS or EDW conditions are to be calculated using a non-linear
time domain rigid body hydrodynamic program with the capabilities to calculate the hydrostatic forces and wave
incident forces (Froude-Krylov forces) taking into account the change in hull shape above and below the waterline.
The time domain simulation can be running using either the EDS or the EDW as described in 3.2 and 3.3. The time
interval in the time domain simulation should not be greater than one tenth of the lowest period of the encounter waves.
For the EDW assessment process, the time domain simulation should be run for at least 20 wave periods. A ramp up
period of two wave periods should be specified and the data for the first say 5 periods worth of data should be ignored
in the post processing in order to allow the response to stabilise.
For the EDS assessment process, the time domain simulation for the non-linear response calculation should be run for at
least 60 minutes. It is to be verified that the data obtained is sufficient to obtain a statistically converged set of data that
is suitable for prediction of the probable maximum value in 3 hours. A ramp up period of around 25 seconds should be
specified and the data for the first say 100 seconds worth of data should be ignored in the post processing in order to
allow the response to stabilise.
However the following approach is recommended:

1. The non-linear rigid body time domain analysis is run for the same length of time as the whipping analysis, see 4.8.
2. It is also highly recommended that a linear rigid body time domain analysis of the same length is also undertaken.
This approach allows the statistical analysis for the linear, non-linear and the whipping responses to be based on exactly
the same wave data. This approach gives the same number of hog and sag events and ensures consistency of approach
with regard to the statistical analysis. The results displayed in Figure 10 and Figure 11 were based on this approach.
3.4.1 Wave Components of the EDS approach
The amplitude of the each wave component making up the selected EDS is to be calculated using the spectral energy of
the EDS. The wave amplitude at each wave frequency a i is given by
a (i ) 2S (i )
The time history of the wave elevation t at location xw is defined as:
Nw
t (xw , t)
i 1
a ( i ) cos( k i x w i t i )
where

MTES/SAH Draft v1.5
S ( ) is the spectral value of EDS
i2
ki is the wave number
g
i phase shift of each wave component. This is to be randomly assigned. Note it is suggested that some pre-
selected sets of random phase angles are used, so that repetition of analyses is possible.
Nw the number of wave components. N w should be at least 100 for a one hour time domain simulation run and
over 250 for a 3 hour run. The number of wave components is to be sufficient to ensure that the time trace of
the wave elevation does not repeat or show similar repetitive patterns throughout the duration of the
simulation. Alternatively several shorter runs can be used, but each simulation is to have a different set of
random phase angles.
wave frequency interval. The first and last wave frequencies are to be chosen such that the following
relationship is satisfied
S (i ) 0.001S max
S max maximum value of S (i )
3.5 Derivation of hogging and sagging non-linear factors for the vertical wave bending moment
3.5.1 Time domain simulation using the EDW approach
Please note the recommended method is to use the EDS approach.
The peaks and troughs (corresponding to the hogging and sagging wave bending moments, assuming a positive value
correspond to a hogging moment) of the time domain trace of non-linear vertical wave bending moments are to be
extracted from the time traces after subtraction of the still water bending moment (SWBM). An example time trace is
shown in Figure 4.
The SWBM to be subtracted is the SWBM calculated by the hydrodynamic code at the start of the time domain analysis
before any wave elevation has been applied. The value from the loading manual is not to be used.
The non-linear hog and sag ratios, equivalent to ffH and ffS defined in the Ship Rules, Pt 4, Ch 4, 2.4, are derived as
follows:
M VS
f fS for sagging, but is not to be taken less than the standard rule value, see Pt 4, Ch 2, 2.4.1 or Pt
M VL
3, Ch 4, 5.2.1 depending on ship type. Note ignore the sign of the sagging correction in the
comparison with the rule value; e.g. if f1 is -1.1, then ffs has to be less than -1.1, hence -1.05 is
not acceptable and -1.25 is.
M VH
f fH for hogging, but is not to be taken less than the standard rule value, see Pt 3, Ch 4, 5.2.1
M VL
where:
MVS is the maximum sagging moment in the time domain simulation.
MVH is the maximum hogging moment in the time domain simulation.

M VL M wo if the Rule linear design bending moment was used to define the EDW, see 3.2,
or

MTES/SAH Draft v1.5
M VL VBM LT if the long term value at the 10-8 level of probability of exceedence, see 2.9, was used to define the
EDW.
3.5.2 Time domain simulation using the EDS approach
The probability distributions of the sagging and hogging bending moment distributions of the time domain simulation
are derived using a peak counting method after subtraction of the SWBM. Both peaks and troughs (corresponding to the
hogging and sagging wave bending moments, based on the usual sign convention of the vertical bending moment) of
the time domain signal should be counted separately as one peak or trough per mean zero crossing period, see Figure 9.
The SWBM to be used is the one that is calculated by the hydrodynamic code and not the value from the loading
manual, as the calculated one is the applicable SWBM for this time domain analysis.
The probable maximum hogging and sagging bending moments in three hours are to be obtained by fitting a 3
parameter Weibull or similar distribution to the probability of exceedence of the peak values. The method to do this is
described in see 5.1.
The final non-linear hog and sag ratios are derived as follows:
3hr
M VS
f fS 3hr
for sagging, but is not to be taken less than the standard rule value, see Pt 4, Ch 2, 2.4.1 or Pt
M VL
3, Ch 4, 5.2.1 depending on ship type
3hr
M VH
f fH 3hr
for hogging, but is not to be taken less than the standard rule value, see Pt 3, Ch 4, 5.2.1
M VL
where:
3h
MVS is the probable maximum sagging bending moment in 3 hours.
3h
MVH is the probable maximum hogging bending moment in 3 hours.
3h
MVL is the probable maximum linear bending moments in 3 hours. This can be derived from short term
statistical analysis using the EDS or from a linear time domain simulation using the same time-train of the
non-linear time simulation but ignoring the non-linear hydrostatic forces and wave incident forces due to the
hull shape.
3.6 Design vertical wave bending moments
The design values of vertical wave bending moments due to non-linear effects of hull shape are to be taken as follows:
VBM NL S f fS M wo for sagging
VBM NL H f f fH M wo for hogging

where
Mw0 rule wave bending moment, see Part 3, Ch 4, 5.2.1. Note: the Rule length distribution factor C2 is to be applied.
3.7 Application to the structural assessment
The design values of vertical wave bending moments due to non-linear effects of hull shape are to be used for the
normal structural assessment in accordance with the Ship Rules. Hence these values replace the Mw rule values in Pt 4,
Ch 2 and are to be applied to the:
Rule inertial and section modulus requirements and local scantling requirements
Rule buckling requirements

MTES/SAH Draft v1.5
SDA procedure
The assigned rule permissible SWBM should be applied, see 2.2.
It is necessary to review all critical locations along the length of the ship. Typically this will include:
Midship section
Forward and aft ends of closed and open cross sections, so in way of the forward end of the closed deck
area of an aft accommodation block and the aft end of the container hold open section just in front of the
accommodation block
Cross sections where there is a significant change in longitudinally effective scantlings

MTES/SAH Draft v1.5
4 Assessment of the hull girder loads including whipping loads
4.1 General
In this section, requirements are given to derive the whipping enhancement factor of the design wave bending moment
using first principles direct calculation procedure. It is also necessary to derive the longitudinal distribution of the wave
bending moment and vertical shear force in order to ensure that the hull girder strength of the ship is valid along its
length.
The process to derive the whipping response of vertical shear force, horizontal bending moment and shear force and
torsional moment is the same as that to derive the whipping vertical bending moment response.
The procedure derives the design wave bending moment based on the Equivalent Design Wave (EDW) or the
Equivalent Design Sea states (EDS) approach. It is only necessary to apply one of these two methods and LR's preferred
method is to apply the EDS approach.
Note
At the moment, whipping and springing analysis methods are under development and all organisations
involved in this topic are still refining procedures, hence this guidance notes presents the best methods
currently considered viable.
The design values are then given by applying this EDW or EDS to a non-linear time domain hydroelastic ship motion
analysis.
The procedure to derive the non-linear response for other load values such as horizontal bending moments and torsional
moments is similar.
Whipping is more susceptible to draft and speed that the extreme wave bending moment and hence the procedure is to
consider different ship operation conditions and different ship speeds. As a consequence it is now necessary to review
several EDWs or EDSs in order to evaluate the design values.
Figure 5 shows part of a time trace from full scale measurements and the time trace from a whipping calculation after a
whipping impact occurred. Note also that the time series of the wave conditions for the measurements is not the same as
for the calculation; hence the two motions traces are only comparable with respect to the trends they show.
Time series of VBM amidships
7.0E+06
measurement
6.0E+06 calculation
5.0E+06
4.0E+06
kNm
3.0E+06
2.0E+06
1.0E+06
0.0E+00 s
0 20 40 60 80 100
Figure 5 Typical time series of VBM amidships
4.2 Determine EDW and additional equivalent design waves
The basic EDW used here is as defined in 3.2

MTES/SAH Draft v1.5
As for the non-linear response, it is necessary to review the sensitivity of response to small variations in encounter wave
period, so extra AEDWs may need to be selected. The designer is to demonstrate that the selection of wave period is
such that the maximum whipping response has been determined.
4.3 Derivation of the EDSs
Several EDSs are needed to derive the whipping enhancement factor of the design vertical wave bending moments to
ensure that large whipping responses are calculated. EDSs are to be selected for each ship operational speed defined in
2.6. Each EDS is defined as the sea state which has the maximum contribution to the 10-8 level of probability of
exceedence in the long term statistical analysis of the vertical wave bending moment amidships at each ship operation
speed. The method specified in 3.3 is to be used, after adjustment for ship speed.
The derivation of the wave components of the EDS approach is given in 3.4.1.
4.4 Loading condition
The loading conditions to apply are specified in 2.2.1.
The mesh size for the hydrodynamic model is specified in sections 2.3 to 2.6.
4.6 Structural model of the ship
The structural flexibility should be taken into account for the derivation of the whipping enhancement factor of the
design wave bending moment by adding the natural mode shapes of ship hull structure to the rigid body modes. These
flexible natural modes are computed using the Finite Element method based on either a 2D model (beam model) or a
3D model of the ship structure. At least the first 5 modes of hull girder vertical vibration should be taken into account in
the analysis.
4.6.1 2D structural modelling
While a 2D FE model is adopted in the modal analysis of ship hull structure, the hull girder should be modelled as a
non-uniform beam according to the so called Timoshenko beam theory and should be modelled using at least 50 beam
elements. Ideally the length of beam element and position should match the hydrodynamic mesh is so far as possible.
The total length of the beam model should match the total length of the ship and will need to be tapered at the ends to
ensure that the hull girder flexibility is correctly modelled.
If a 3D FE model is available, but not being used for the whipping/springing analysis, then the principal hull girder
bending (and torsional) dry mode eigenvectors of the 2D model should give similar results to the 3D dry mode
eigenvectors.
The calculation of the hull girder inertia and section modulus is to be done in accordance with Pt 3, Ch 3, 3.4, which
defines the material that is longitudinally effective and the shadow areas. It is to be assumed that all hull girder section
properties transform smoothly along the length of the ship and hence there are no abrupt discontinuities in the properties
due to changes in cross section between closed sections and sections with large deck openings. This applies to all hull
girder properties including the shear centre for torsional moment.

MTES/SAH Draft v1.5
4.6.2 3D structural modelling
When a 3D FE model is adopted for the modal analysis of ship, the ship's structure should be modelled using shell
elements and bar elements to represent the whole ship. This model should extend over the full breadth and depth of the
ship and represent, with reasonable accuracy, the actual geometric shape of the hull.
All primary longitudinal structures (i.e. all longitudinal bulkheads, decks, stringers) are to be modelled using shell
elements. Similarly, all transverse primary structures (i.e. watertight bulkheads, open bulkheads (mid-hold support
structure), web frames and cross-deck structures) are to be represented in the model.
The size and type of shell elements selected are to provide a satisfactory representation of the deflections and stress
distributions within the ships structure. In general, the mesh of the shell element is to follow the primary stiffening
arrangement. Hence, it is anticipated that there will be:
transversely, one element between longitudinal girders;
longitudinally, one element between double bottom floors; and
vertically, one element between stringers or decks.
The ships superstructure or deckhouse is to be included in the model. This is to be represented using shell elements
with a mesh arrangement similar to that used for the hull and which adequately represent the overall structural
arrangement of the deckhouse.
Secondary stiffening members and effective longitudinals may be modelled using line (bar) elements, grouped at plate
boundaries, positioned in the plane of the plating having an axial property with the cross-sectional area representing the
stiffener area or may be ignored but their sectional areas are to be added to the thicknesses of the plate elements.
4.6.3 Eigenvector analysis techniques
Special care is to be taken to ensure that the structural idealisation of the FE model does not introduce vibrational
responses (natural modes) which are not realistic or are not applicable for the assessment of global hull responses. In
some cases, additional artificial constraints may need to be introduced to ensure that these unnecessary modes are
suppressed.
Except the structural mass, other mass from dead weight may be placed at intersections of primary longitudinal
structures and primary transverse structures to avoid unnecessary local vibration modes.
The free - free boundary condition should be used for the modal (eigenvector) analysis.
It is acceptable to use dynamic or static reduction techniques for the 3D whole ship model to:
remove unwanted vibration modes of localised structural responses
remove spurious vibration modes due to localised poor structural modelling in areas of no interest
reduce the number of possible vibration modes to be analysed
This is applicable to even the coarse mesh FE model specified above.
Fore detailed whole ship FE models, it is recommended that, as a minimum, the following dynamic reduction node set
(the A-Set in NASTRAN terms) is applied:
At key points of the cross section at each transverse bulkhead
At key points of the cross section at each mid hold or compartment length
The above recommendation is applicable for a whipping analysis that is concentrating on the calculation of the global
hull responses. If a more detailed analysis of a structural area is required, then it is necessary to add the following node
points to the dynamic reduction node set:
At all intersections of primary members in way of all areas of interest
The free - free boundary condition should be used in the modal analysis.

MTES/SAH Draft v1.5
Typically the eigenvector analysis should assess at least the first 50 modes. The whipping analysis is to select all major
hull girder vibration modes from this set. For assessment of vertical bending whipping responses, the first four vertical
hull girder vibration modes should be included as a minimum requirement.
The more modes included the better the resulting stress solution, so more modes that those required for the springing
analysis have to be included, eg if only springing at the first 4 modal shapes is required, the stress analysis will require
10 to 14 modes to get an accurate stress results. Using the first 4 modes will give very poor stress predictions. The
original eigenvector analysis and dynamic reduction node set have to be suitable for the locations where stresses are
required. Note that it important to ask for 50 eigenvector modes to be calculated and discard 40 of these than ask for the
20 modes and discard 10 - this will ensure that the best lower frequency modal shapes will be captured and important
modes will not be missed. For localised stresses, then many more modes may need to be calculated
4.6.4 Loading a whole ship 3D FE model
It is necessary to apply all the lightship and deadweight items as structural element mass items (QUADS, BARS, etc) or
as MASS elements. It is important to ensure that the required displacement, , longitudinal centre of gravity, LCG, and
vertical centre of gravities, VCG, and roll radius of gyration, kr, are achieved in the whole ship FE model. If only
vertical bending responses are required, then matching the correct VCG is not essential. However, if horizontal bending
and torsion response is required, then matching the loading condition VCG and kr is critical.
Liquids in tanks need to be converted to nodal masses around the tank boundary that give:
1. The correct mass centroid properties for the tank
2. The correct mass inertia. For small tanks, this latter requirement is not that critical. For large tanks then it is critical.
In this case, use of nodal masses around the tank boundary is not adequate. It is then necessary to model the liquid
as mass items that are linked to the tank boundary using NASTRAN RBE3 elements or similar in a similar to that
used to define container stacks.
Containers in stacks need to be modelled in such a way that the vertical and transverse distributions of container masses
are accurately modelled. Failure to do this will mean that the required loading condition VCG and roll radii of gyration
are not maintained. The recommended way of doing this is as follows:
1. Model each container stack as a separate mass item
2. Link this mass item to the relevant part(s) of the structure via the use of NASTRAN RBE3 elements (or
equivalent in other FE packages), see Figure 6.
Hence for containers on deck, define a MASS item for a group of container stacks link each MASS item to the
hatch cover at its forward and aft corners by at RBE3 elements to every container corner on the hatch coaming.
Similarly for containers in holds, for these it may be necessary to represent the attachment to the stack guides using
RBE3 elements. Figure 6 shows an example of container modelling, here the deck containers have been split into 4
groups with the Mass element (in cyan) at the VCG of the container stack. Each mass is then linked by an RBE3
element to the nearest point on the hatch coaming to a corner post location. This figure also shows the containers in
the hold similarly modelled and also the hatch covers modelled in a similar fashion.

MTES/SAH Draft v1.5
Figure 6 Example of modelling containers figure will be improved. Ideally each 40ft Bay should be modelled
4.6.5 Structural damping
There are many uncertainties regarding the structural damping and it is not easy to predict or measure. The modal
damping coefficient of the first ship hull vertical vibration mode is typically between 2% and 3% depending on the ship
type and size. Generally the damping ratio grows with increasing natural frequency.
It is recommended that the following formula is used to estimate modal damping coefficients for the whipping and
springing analyses:
i 0.0005 si 0.020
Where
i modal damping coefficient for the ith hull girder natural frequency
i frequency of the ith hull girder natural vibration mode, in rad/s
If possible the values of structural damping are to be derived on the basis of achieving a comparable springing response
to that obtained for a similar reference ship for which full scale measurements are available.
4.7 Hydrodynamic definition of the bow flare and slamming regions
4.7.1 Impact load evaluation
For container ships the main factor affecting the whipping responses of the hull girder is bow flare slamming due to
their highly flared bow shape. For other ship types, then bottom slamming would be important if the draught is low.
Stern slamming is also an issue for container ships but is not usually as important as bow flare shape.

MTES/SAH Draft v1.5
The wave impact forces due bow flare impacts, bottom slamming or stern slamming are to derived by using software
linked to the time domain non-linear ship motion code. Suitable methods include:
Momentum slamming theory
Generalised Wagner theory
VOF CFD codes (Volume of Fluid) Computational Fluid Dynamics
In general only 2D methods are applied due to computational issues with 3D approaches. Methods such as Generalised
Wagner theory or Momentum slamming theory are usually restricted to solving impact loads for bow flare shapes only,
in addition they also require that cross sections must have increasing waterline offsets, hence they can't model bulbous
bows. In this case the decreasing waterline offsets should be ignored; hence the resulting bow shape is modified to have
a vertical side. This issue also means that these methods fail for large roll angles and hence calculating the whipping
response in oblique seas can be problematic.
4.7.2 Definition of the bow flare impact sections
The region forward of the parallel midbody with significant bow flare, typically forward of 0.8L, is to be defined for the
bow flare impact calculation. At least 15 2D transverse cross-sections are to be generated over the forward region.
These 2D cross-sections are to be inclined to the vertical in such a way that the water flow over the hull will be parallel
to the cross-section plane, see Figure 7. Typically an angle of around 30 degrees will need to be applied.
The following angle to the vertical is suggested as a guideline

degrees
where
Rhv = a c 0.515
Rvv = 0.5
a = 10
= 0.14
.
e = 1
Lc = approximate critical wavelength in m, see 3.2

Tc = seconds
c =
V = ship speed in knots
Figure 7 Example of cross-sections defined for bow flare slamming
4.8 Calculation of whipping vertical bending moments in time domain
The whipping responses of the vertical wave bending moments in time domain are to be calculated using a time domain
hydroelastic hydrodynamic program with capabilities to deal with the non-linear hydrostatic forces and wave incident
forces (Froude-Krylov forces) due to the hull shape as well as the impact forces.

MTES/SAH Draft v1.5
The time domain simulation can be run using either the EDW approach or the EDS approach described in 4.2 and 4.3.
The time step in the time domain simulation should not be greater than one tenth of the lowest period of the dynamical
system in the time domain simulation which considers both encounter waves and ship vibrational modes. Hence if the
highest natural mode of vibration (e.g., the 5th mode) in the simulation has a (wet mode) frequency of 2 Hz, then the
time step should be less than 0.05 s (20 Hz).
For the EDW assessment process, the time domain simulation should be run for at least 15 wave periods. A ramp up
period should be specified and the data for the first few periods worth of data should be ignored in the post processing
in order to allow the response to stabilise.
For the EDS assessment process, the time domain simulations for the whipping response calculation should be run for
at least 180 minutes and preferably 300 minutes in order to be able to predict of the probable maximum value in 3 hours
based on a statistically converged set of whipping response data. This can be achieved by one run or by several shorter
runs of say 60 or 30 minutes and combining the results if necessary, see also 3.4.
The whipping response is a very stochastic process and as a consequence it is necessary to ensure that the whipping
bending moment response has converged.
Notes
Convergence studies have shown that there is still a significant variation of probable maximum in 3 hours for
a total simulation time of 180 minutes. Full convergence was only achieved after a total simulation time of 300
minutes. See Figure 8
Figure 8 Example of convergence of whipping hog factor with duration of simulation time
4.9 Derivation of hogging and sagging whipping factor for the vertical wave bending moment
4.9.1 Time domain simulation using the EDW approach
The peaks and troughs of the vertical wave bending moment time domain signals for each EDW and AEDW are to be
identified after subtraction of the SWBM.

MTES/SAH Draft v1.5
The SWBM to be used is the one that is calculated by the hydrodynamic code and not the value from the loading
manual, as the calculated one is the applicable SWBM for this time domain analysis.
The whipping enhancement factors, similar to ffH and ffS defined in Pt 4, Ch 4, 2.4, are to be calculated for the EDW and
for each AEDW. The final value of the hog and sag whipping enhancement factors is to be taken as the maximum of all
values calculated.
M WS
f fS W for sagging, but is not to be taken less than the non-linear calculated value of ffS
M VL
M WH
f fH W for hogging, but is not to be taken less than the non -linear calculated Rule value of ffH
M VL
where:
M WS is the mean of the largest 5 maximum sagging moments, including non-linear whipping effects, observed
during each individual EDW or AEDWs time domain simulation being considered.
MWH is the mean of the largest 5 maximum hogging moments, including non-linear whipping effects, observed
during each individual EDW or AEDWs time domain simulation being considered.
M VL M wo if the Rule linear design bending moment was used to define the EDW, see 3.2,
or
M VL VBM LT if the long term value at the 10-8 level of probability of exceedence, see 2.9, was used to define the
EDW.
4.9.2 Time domain simulation using the EDS approach
The probability distributions of the sagging and hogging bending moment distributions of the time domain simulation
are derived in a similar manner to that described in 3.5.2 for each EDS.
However, the number of peak and trough values of the bending moment signals of a flexible ship analysis are far higher
than those of a rigid body analysis. Therefore the peak and trough values should be recorded on the basis of one per
mean zero crossing period of the equivalent rigid-body time domain signal. This can be achieved by either:
Apply this counting approach to a non-linear rigid body time domain analysis run using the same wave-train,
see 3.4. This is the recommended method.
Or apply a low-pass filter on the signal of the flexible ship can be applied to obtain the approximate signal for
a rigid body, as long as the natural vibrational frequencies are much higher than wave frequencies and wont
be filtered out. This approximate rigid body signal can then be used to define the non-linear response in Figure
9 and hence allow extraction of the peaks and troughs for the whipping response.

MTES/SAH Draft v1.5
Example of counting for extremes
1.0E+07
5.0E+06
Linear
VBM (kNm)
0.0E+00 Linear count

Nonlinear
Nonlinear count
-5.0E+06 Whipping
Whipping count
-1.0E+07
-1.5E+07
150 160 170 180 190 200
Time (s)
Figure 9 Counting method for extremes of the linear, non-linear and whipping response time domain traces
The maximum expected hogging and sagging bending moments which include the whipping effects in three hours are
obtained by fitting a 3 parameter Weibull or a 3 parameter log-normal or similar distribution to the probability
distribution of the peak values, see 5.1. Examples of weibull curve fitting and the derivation of the probable maximum 3
hour values are shown in Figure 10 and Figure 11.
The whipping factor is to be taken as follows:

3h
M WS
f fS W 3h
for sagging, but is not to be taken less than the non-linear calculated value of ffS
M VL
3h
M WH
f fH W 3h
for hogging, but is not to be taken less than the non-linear calculated value of ffH
M VL
where:
3h
M WS is the maximum value of the probable maximum whipping sagging moment in 3 hours from the time
domain simulations for each EDS.
3h
M SH is the maximum value of the probable maximum whipping hogging moment in 3 hours from the time
domain simulations for each EDSs.
3h
MVL is the maximum value of the probable maximum linear bending moment in 3 hours for each EDS. This can
be derived from short term statistical analysis using the EDS, as shown in 3.3 or from a linear time domain
simulation using the same EDS but ignoring the non-linear hydrostatic forces and wave incident forces due to
the hull shape as well as the impact forces.

MTES/SAH Draft v1.5
Figure 10 Weibull fitting of the linear, non-linear and whipping vertical sagging bending moment distributions
Figure 11 Weibull fitting of the linear, non-linear and whipping vertical hogging bending moment distributions
4.10 Design vertical wave bending moments
The design values of vertical wave bending moments due to non-linear effects of hull shape and whipping are to be
taken as follows:

MTES/SAH Draft v1.5
VBM NL S f fS W M w 0 for sagging

VBM NL H f fH W M w 0 for hogging
where
Mw0 Rule wave bending moment, see Part 3, Ch 4, 5.2.1.
Note: the Rule length distribution factor C2 is to be applied in the Mw0 equation
4.11 Application to the structural assessment
The design values of vertical wave bending moments due to non-linear effects of hull shape and whipping are to be
used for structural assessment as follows:
Rule buckling requirements for stiffeners. For this application the buckling factor in Lloyds Registers Ship
Rules, Pt 3, Ch 4, 7.5 is be taken as 1.0 for longitudinals.
Ultimate strength assessment based on the same approach as that adopted by the IACS Common Structural
Rules for Double Hull Oil Tankers (CSR-DHOT) 2012.
It is necessary to review all critical locations along the length of the ship. Typically this will include:
Midship section
Forward and aft ends of closed and open cross sections, so in way of the forward end of the closed deck area of
an aft accommodation block and the aft end of the container hold open section just in from of teh
accommodation block
Cross sections where there is a significant change in longitudinally effective scantlings
4.11.1 Ultimate strength assessment
The ultimate strength of the hull girder at all critical locations along the length of the ship are to be evaluated. The
incremental-iterative method specified in the Common Structural Rules 2013 or the Harmonised Common Structural
Rules may be adopted. Alternatively the LR procedure encompassed within the RulesCalc software based on LRPASS
software LR20202 and LR20203 or equivalent procedures may be adopted.
The criteria to apply is

M US
M SS W VBM NL S for sagging
R
M UH
M SH W VBM NL H for hogging
R
Hence the ultimate Load Utilisation Factor LUF must be less than one
LUFSag
M SS W VBM NLS 1 for sagging
M US
R
LUFHog
M SH W VBM NL H
1 for hogging
M UH
R
where
MSS maximum sagging SWBM that is practical at the draft being considered, see 2.2
MSH maximum hogging SWBM that is practical at the draft being considered, see 2.2
MUS ultimate bending capacity in sagging based on deducting one half the standard deduction for corrosion, dt given
in the Ship Rules, Pt 3, Ch 4, Table 4.7.1 from all structural members

MTES/SAH Draft v1.5
MUH ultimate bending capacity in hogging based on deducting one half the standard deduction for corrosion, dt
given in the Ship Rules, Pt 3, Ch 4, Table 4.7.1 from all structural members
W partial factor for wave bending moment, to be taken as 1.2

R partial factor for capacity, to be taken as 1.1
Notes
1. The single step method in the CSR- DHOT is not applicable.
2. LR20203 and LR20202 are included as the ultimate strength procedure in Lloyds Registerss RulesCalc software.
Otherwise these are part of the LRPASS software.
3. The CSR requirement is based on tnet50 thickness values. On this basis an allowance of 50% of the standard
corrosion for buckling in the Ship Rules is to be made in deriving the ultimate strength.

MTES/SAH Draft v1.5
5 Statistical post processing
5.1 Evaluation of the probable maximum value in 3 hours
The probable maximum values in three hours are to be obtained by fitting a 3 parameter Weibull or similar distribution
to the probability of exceedence of the peak values.
The Weibull curve fitting routine should be chosen such that it concentrates on a best fit over the tail of the distribution.
It is recommended that the fitting distribution should apply equal weighting across the distribution and should not take
account of the number of values within each probability bin. As the best fit over the tail of the distribution is required,
then it is recommended that the first few probability bins are excluded from the curve fitting process to improve the fit
over the tail. See Figure 12. Typically the lowest 20% of the data may be ignored, but the exact value needs to be
identified through trial and error to minimise the error ratio.
This method ensures that the fitting process concentrates on the tail of the distribution. This is illustrated in Figure 12,
a) shows an example of fitting based on the distribution being weighted by the number of responses in each bin and
using all data points; b) shows an example of fitting based on each response bin being given equal weighting and the
lowest response values ignored.
Well fitted Weibull plots are shown in Figure 10 and Figure 11.
Fitted Weibull cumulative distribution_whipping-sag Fitted Weibull cumulative distribution_whipping-sag
1
1
Q fit curve
Q fit curve
0.1
Cumulative probablity
Cumulative probablity
0.1
0.01 0.01
0.001 0.001
0.0001 0.0001
0.0E+00 5.0E+06 1.0E+07 1.5E+07 2.0E+07 2.5E+07 3.0E+07 0.0E+00 5.0E+06 1.0E+07 1.5E+07 2.0E+07 2.5E+07 3.0E+07
VBM (kNm) VBM (kNm)
a) Poor fitting Weibull cumulative distribution. Fitting b) Good fitting Weibull cumulative distribution. Fitting
used the whole data set for whipping BM responses based on using the tail of whipping BM responses
Figure 12 Example of good and poor fitting of Weibull distributions
A suitable curve fitting routine is as follows:

The best fit is obtained by minimising the Error Ratio, ER, given by:
ln lnQ
Nb
ER a n ln Q f 2
n 1
Where
Nb number of probability bins ignoring the excluded bins
Qa actual cumulative probability value associated with bin n
Qf cumulative probability value of the bending moment assigned to bin n based as derived using the
curve fit parameters.
Typically for a 3 parameter Weibull distribution, it is suggested that the location parameter, , is adjusted incrementally
and the shape, , and scale, , parameters derived from the curve fit. The ER value is derived for each value of location

MTES/SAH Draft v1.5
parameter until the ER is minimised. Note that ER is not a true standard deviation parameter but is useful to demonstrate
best fit for this application.
3 parameter Weibull distribution parameters

X

Cumulative probability, Q Q e
Probable maximum value Mf for a cumulative probability of Qa

M f lnQa
1

It is recommended that a value of ER of less than 0.01 is achieved if possible, however not more that 20% of the smaller
data values should be excluded from the assessment (excluded probability bins). When deriving the probable maximum
values from the Weibull distributions of linear, non-linear and whipping responses, similar error values and goodness of
fit criteria should be applied to all fitting processes for consistency.
It should be noted that the final non-linear hog and sag ratios, see 3.5.2, and the whipping hog and sag ratios, see 4.9.2,
are very strongly affected by the curve fitting process and hence great care and consistency of approach is required. One
of the critical issues is the amount of data available and the longer the duration of the time simulation the more data is
available, see 3.4.1 and 4.8. As noted earlier whipping is a stochastic process and hence it is likely that the tail of the
distribution will be sparsely populated, hence the results are heavily affected by the number of data points.

MTES/SAH Draft v1.5
6 Assessment of fatigue including hull girder springing loads
6.1 General
In this section, requirements are given to calculate the effects that springing has on the predicted fatigue life. Ideally the
effects of springing should be directly included in the spectral fatigue calculation but this is very complex when the
non-linearities of the dynamic load response, eg intermittent wetting of the hull in way of the static waterline or non-
linear ship motion response in large waves, are also included. In view of this the proposed method derives a springing
fatigue reduction factor based on a linear spectral fatigue assessment procedure. This springing fatigue reduction factor
can then be applied to the non-linear "rigid body" spectral fatigue calculation to obtain a predicted fatigue life including
springing effects.
The first part of the process is to determine the springing response, this is followed by a fatigue assessment using the
stress RAOs including the springing responses and also a fatigue assessment excluding the springing responses. The
ratio of these two fatigue lives give the springing fatigue reduction factor that is applicable for the structural element
being considered.
The occurrence of springing of a ship can be reliably predicted by hydroelastic ship motion programs, however the
exact determination of the magnitude of springing response is more difficult. Hence it is necessary to calibrate the
hydroelastic analysis method based on full scale measurements of a reference ship.
Figure 13 shows the calculated spectral energy response in the wave frequency range for vertical bending moment of an
8500TEU container ship against the measured response from full scale measurements. As can be seen, the comparison
is very good. Some whipping response was observed in the full scale measurements, however it should be noted that the
spectral energy of whipping responses will normally be very small, so what is being shown here is mainly the springing
response around a wave frequency of 3.5 rad/s.
VBM spectrum amidships

6.0E+05
measurement
5.0E+05 calculation
4.0E+05
kNm
3.0E+05
2.0E+05
1.0E+05
0.0E+00
0 1 2 3 4 5 6
wave frequency (rad/s)
Figure 13 Measured and calculated vertical bending moment amidships for a JONSWAP wave spectrum
The following process does not include any allowances for whipping induced fatigue. Whipping responses will result in
a decrease in the fatigue life, but it is not thought that whipping is a major contributor to fatigue due to the relatively
low number of cycles associated with whipping during a ship's lifetime.
6.2 Outline of Assessment Procedure
Several methods for the determination of the springing response for inclusion in the fatigue assessment are given in this
document. The reason for inclusion of more than one method is due to the fact that this technology is still maturing,

MTES/SAH Draft v1.5
there are several valid ways to approach the problem and industry is in the process of establishing the most suitable
methods for the calculation of springing and the assessment of fatigue including springing.
The proposed methods are:

1. Assessment of linear stress RAOs for all headings and speeds including springing responses
2. Assessment of linear bending moment RAOs for all headings and speeds including springing responses
3. Assessment of non-linear springing responses for critical wave frequencies
The aim of these assessment procedures is to allow a "Springing Fatigue Reduction Factor" to be derived, see 6.3. This
can be used to adjust the results of the Level 3 Fatigue Design Assessment procedure (FDA3) to allow for springing
responses.
The choice of which springing response method to use is to some extent dependant on the chosen fatigue assessment
process and the available hydroelastic and fatigue assessment tools.
6.2.1 Assessment of linear stress RAOs for all headings and speeds including springing responses
The linear stress RAO approach is the principal method currently used by Lloyd's Register. This approach uses a
frequency domain hydroelastic analysis based on a full ship 3D FE model. It allows stress RAOs to be derived on any
element of the FE model taking into account all external hydrodynamic and hydroelastic pressures and internal inertial
loads; thereby accounting for all local, primary and global hull girder load actions including vertical bending and shear,
horizontal bending and shear and torsional and warping effects.
Determination of the stress RAOs including springing actions for any element of the full ship FE model allows a full
spectral fatigue assessment to be undertaken using spectral techniques suitable for bi-modal spectral response to derive
the fatigue damage in each sea state.
The linear stress RAO approach is fully supported by Lloyd's Register's software tools as follows:
HydroModeller Creates the hydrodynamic mesh from offsets, FE models or similar.
WAVELOAD-FD Linear rigid body frequency domain 3D diffraction ship motion code
HydroE-FD Linear flexible body frequency domain 3D diffraction hydroelastic ship motion code
The linear hydroelasticity approach is described in more detail in 6.4.2.
The stress RAOs can be calculated by back substituting the calculated magnitudes of the principle modes from the
hydroelastic analysis into the finite element eigenvector analysis. These stress RAOs are exact in the sense that they
include all load actions and structural responses, so include the response due to local inertia and dynamic pressure loads.
The stress RAO of any element i in the structure can be represented by below formula;
n
StressRAOi SEi,k RAOk
k 7
Where,
k = 1 to 6 are 6 rigid body modes, 7 to n are the flexible body modes
SEi ,k = element stress eigenvector of flexible mode k
RAOk = the RAOs of the each flexible mode k (complex value with different magnitude and phase angle for
each mode)
Figure 14 shows the stress RAOs calculated by the process of back substituting the magnitudes of the principle modes
into the FE eigenvector analysis, labelled as "calculated by stress modal shapes". These are compared with the stress
RAOs calculated by using the keel section modulus and the elastic bending moment response and 2D beam theory
"calculated by HydroE-VBM". As can be seen these curves are very similar for this case as it is a head sea case and 2D

MTES/SAH Draft v1.5
beam theory is adequate for this keel location. Please note that this ship was a very flexible body with a low natural
period of vibration. The hull girder was so flexible that the elastic response to long waves (wave frequency of 0.4 rad/s)
was significantly reduced compared to the "rigid" body response, see the "calculated by rigid body-VBM" curve.
Figure 14 Stress RAO including springing effects for a very flexible ship
In order to derive accurate stress RAOs, sufficient flexible modes have to be included in the hydroelastic analysis.
Failure to include sufficient modes will result in the stress RAOs being under predicted. Typically at least 8 modes
need to be included in the hydroelasticity calculation, but a check similar to that shown in Figure 14 may be necessary
to confirm this. Failure to include a sufficient number of flexible modes will result in the stress RAO being significantly
underestimated. It is also necessary to ensure that the chosen dynamic reduction node set is sufficient to allow the
stresses to be correctly predicted for the locations being reviewed, see 4.6.3.
Using this approach, it is possible to derive stress RAOs including springing effects for any element within a coarse
mesh FE model, provided that the chosen dynamic reduction node set is sufficient. Hence locations close to hatch
corners can be assessed taking into account global longitudinal and horizontal bending and shear actions and torsional
actions.
6.2.2 Assessment of linear bending moment RAOs for all headings and speeds including springing responses
It is also possible to use a linear bending moment RAO approach. This approach uses a frequency domain hydroelastic
analysis applied to either a full ship 3D FE model or a 2D beam element FE model of the ship. It allows the global
bending moment RAOs to be derived along the ship taking into account all external pressures and internal inertial loads
including those associated with hydroelastic issues.
Determination of the bending moment RAOs including springing actions at any longitudinal location of the ship allows
the stress to be derived at selected locations using simple beam analysis techniques. A full spectral fatigue assessment is
then undertaken using suitable bi-modal spectral techniques to derive the fatigue damage in each sea state or other

MTES/SAH Draft v1.5
suitable techniques. This method is limited by the simple beam element calculation of the stress at a location, for
example warping stresses and primary structural responses are not correctly included.
The linear bending moment RAO approach is fully supported by Lloyd's Register developed software tools.
The linear hydroelasticity approach is described in more detail in 6.4.2.
6.2.3 Assessment of non-linear springing responses for critical wave frequencies
The alternative non-linear approaches are based on the assumption that springing of the two node vertical bending mode
is the most critical and the estimation of fatigue life is most critical with respect to vertical bending. Analysis of the
fatigue damage including springing effects for containers ships analysed to date fully supports this assumption. These
approaches are also applicable to torsional and horizontal bending vibration modes, but this becomes very complex.
Typically the method will use a time domain hydroelastic analysis using a 2D beam element FE model to represent the
ship.
The derivation of the global bending moment including springing response at several longitudinal positions will be
necessary to allow the fatigue assessment to be made along the ship.
This method is based on the following premises:

The springing response is dominated by certain critical wave frequencies.
The total springing response can be calculated by adding the contribution from each of these critical wave
frequencies.
The contribution to springing of each critical wave frequency is proportional to the wave energy present at that
frequency and also to the springing response to a unit wave of that frequency.
The wave encounter shift method can be used to derive springing responses for other speeds and sea states.
This allows the springing response to be simply calculated in any sea state and hence assessment of the fatigue life
using a spectral fatigue process which uses suitable bi-modal spectral techniques.
This method allows the effects of springing of the vertical bending modes to be explicitly calculated within the spectral
fatigue analysis in a very efficient way.
The method is described in more detail in 7.
6.3 Springing Fatigue Reduction (SFR) factor
6.3.1 Overview
The traditional spectral fatigue assessment process adopts a quasi-static approach to calculate the element stresses by
applying hydrodynamic pressures and inertial loads to the static FE model and this approach allows any location in the
structure to be assessed in the whole frequency band, this is the basis LR's FDA3 and FDA2 procedures. These
procedures calculate the probability of a ship encountering a particular sea state at a particular speed, heading and
loading condition. The statistical stress responses are then obtained for each sea state, with corrections for non-linear
effects such as intermittent wetting, and the fatigue damage value in that sea state/heading/speed/loading condition
combination are evaluated. The final fatigue damage value is then the summation of the fatigue damage value for all sea
states. Currently these procedures only allow rigid body stress RAOs to be applied and hence this means that dynamic
effects near structural resonant frequencies cannot be considered.

MTES/SAH Draft v1.5
The above process needs to be applied for the stress RAOs including springing effects. Ideally the FDA3 or FDA2
procedures and software should be updated to include this. However inclusion of the non-linear effects as well as
springing effects is very complex and hence an alternative approach is suggested.
A process similar to the standard spectral fatigue assessment is undertaken based on the stress RAO including springing,
so effectively a simplified FDA3 or FDA2 calculation.
A two stage fatigue calculation is undertaken, the first stage determines the fatigue damage including springing effects
for each structural element and the second stage repeats the calculation excluding the springing effects. The ratio of
these two fatigue lives give the springing fatigue reduction factor that is applicable for the structural element being
considered.
The resulting springing fatigue reduction factor can then be applied to the results of the standard "rigid body" (no
elasticity) fatigue assessment results to obtain a fatigue life including springing effects.
The calculation of the springing reduction factor is supported by a Lloyd's Register spreadsheet which may be available
on request.
6.3.2 Calculation of fatigue damage index including and excluding springing for one sea state
The fatigue damage including springing effects for each structural element and for each sea state, ship speed and
heading combination is to be calculated using the spectral techniques adapted for wide band spectra in association with
the stress RAO including springing. The total fatigue damage is the summation of all these sea state, ship speed and
heading combinations.
Typically the fatigue damage is calculated based on several loading conditions, however as this is a comparative
approach, then the calculation the fatigue damage on the basis of the loading condition most likely to experience
springing will simplify the analysis.
A typical stress RAO curve is shown in Figure 15. As can be seen in Figure 15, the stress RAO is a wideband signal
with several peaks.
The stress response energy of an element i at a given ship speed and wave heading in a single sea state is:
RS w ( ) i S ( ) StressRAO ( ) i 2 d

where
S() Wave energy of the sea state being considered
To calculate the corresponding fatigue damage excluding springing effects in a manner that is consistent with the
hydroelastic springing ship motion analysis, it is necessary to calculate the equivalent "rigid body" fatigue damage. This
is achieved by calculating the spectral stress response energy up to the frequency where springing effects are observed.
This is denoted as the springing cut off frequency and is the red line in Figure 15. The same fatigue calculation process
is then applied to the "cut-off" stress RAO response for each sea state, heading, speed combination and the total fatigue
damage is the summation of all these combinations.
Above this frequency it has been assumed that will be the springing induced stresses. The cut-off frequency can be
taken as follows:
cut off 1 2,51
Where
1 = the first major contributing natural frequency of the hull girder
= modal damping coefficient of the first vibration mode of the hull girder

MTES/SAH Draft v1.5
The first major contributing natural frequency of the hull girder can usually be taken as the two node vertical bending
modal value. For this ship, it can be seen that the horizontal and torsional mode responses are not significant for the
stress RAO results, see Figure 16.
Figure 15 Typical stress RAO curve and cut-off frequency, Head sea case showing the dominant vertical bending mode
Figure 16 Typical stress RAO curve and cut-off frequency, Oblique sea case showing the dominant vertical bending
mode and small torsional or horizontal vibrational modes
6.3.3 Fine mesh details and Stress Concentration Factors (SCF)
Typically a coarse mesh FE model will be used for the hydroelastic analysis. In this case, its not possible to assess the
affect of localised fine mesh regions on localised stresses. It is possible to include suitable SCF's into the springing
fatigue analysis but because a comparative approach is being adopted, then this is not that critical.
6.3.4 Calculation of the final fatigue life and damage index
The total fatigue damage index including springing effect is given by:
D
Dt w FDA3
SRF
where
Dt = total fatigue damage value with springing

MTES/SAH Draft v1.5
Dw FDA3 = rigid body fatigue damage value without springing (FDA Level 3 result)
SRF = springing reduction factor
Dw HydroE
SRF = springing induced damage factor
Dt HydroE
Dw HydroE = rigid body fatigue damage value without springing from the hydroelastic analysis
Dt HydroE = total fatigue damage value including springing from the hydroelastic analysis
The resultant fatigue life will be

1 years or this can be expressed as FLtotal SRF . FL _ FDA3
FLtotal
Dt
Where the rigid body FDA level 3 calculated fatigue life is given by:
1
FL _ FDA3
Dw FDA3
It should be noted that if the rigid body FDA level 3 calculated fatigue life is 40 years, a springing reduction factor of
50% will give a total fatigue life of 20 years including springing.
Figure 17 illustrates the effect of heading, wave height and period and ship speed on the contribution of springing
effects on fatigue for a deck plating element near a hatch corner for a large container ship. Each graph shows the total
percentage contribution to the overall fatigue damage value based on summation of the individual fatigue damage
values for the specified heading, wave height, wave period or ship speed. For this example, this clearly illustrates that
springing fatigue issues are dominant at high ship speed, low wave periods and low wave heights. Also the critical
headings that induce significant springing fatigue damage are close to head seas.

MTES/SAH Draft v1.5
Figure 17 Example of the effect of heading, wave height and period and ship speed on the contribution of springing
effects on fatigue.
6.4 Calculation of springing responses
6.4.1 Hydroelastic methods
Springing responses can be estimated using either linear frequency domain or time domain hydroelastic methods.
The linear frequency domain method gives the response amplitude operator RAO curves including the springing
response. This method is an extension of the normal ship motion theory to include elastic body (hydroelastic) effects.
The advantage of frequency domain hydroelastic method is that it can calculate the motion and load responses including
springing actions for all ship speeds and heading in one analysis.

MTES/SAH Draft v1.5
The time domain method can only assess the springing response in a particular sea state for one heading or the
springing response for one wave frequency, one heading and one speed. The time domain method can also allow non-
linear effects of the hull form to be included in the springing response. The time domain method can be used to derive
the RAO curve including springing response, but this is fairly inefficient. More importantly time domain methods allow
the assessment of sub-harmonics or super-harmonics of the springing response which are not available from most linear
frequency domain codes. This is discussed further in 7.1.
6.4.2 Springing responses using linear frequency domain analysis
The springing responses can be calculated using a frequency domain hydroelasticity program. This is applicable for the
fatigue calculation using the springing RAO method in 6.2.1 and 6.2.2. The basis of this method is similar to that given
for the linear design load calculations described in Section 2. Both rigid and flexible vertical wave bending moment
responses (RAOs) to unit regular waves are calculated using a frequency domain hydroelasticity program.
The hydroelastic analysis is to cover the frequency range appropriate for the speed of the ship and the required hull
girder natural vibration frequencies that need to be considered.
The choice of wave height is to reflect the wave frequency being considered. For wave frequencies in the main wave
energy spectrum region, then it is recommended that unit regular wave amplitude is applied, for higher frequencies then
of the chosen wave height should ensure that a wave slope (H/) of 1/10 is not exceeded.
The total springing response in a sea state is carried out using standard short term statistical methods that are adapted to
calculation of multi peak response spectrum.
From this, a fatigue calculation can be undertaken using the rainflow counting method or by utilising a multi peak short
term spectral calculation, such as specified in the following papers:
1. A new fatigue prediction model for marine structures subject to wide band stress process. J B Park, J Choung,
K S Kim, Ocean Engineering 76 (2014) 144151.
2. Estimation of stress range distribution due to wide banded random loading obtained by rain-flow counting
method, J B Park, K S Kim & J Choung, Advances in Marine Structures. 2011.
3. Dirlik ref?
6.5 Loading condition
The loading conditions to apply are specified in 2.2.2.
The mesh size for the hydrodynamic model is specified in sections 2.3 to 2.6.
6.7 Structural model of the ship
This is taken as specified for the whipping assessment, see 4.6.
6.8 Structural damping
This is taken as specified for the whipping assessment, see 4.6.5.
6.9 Calibration of springing responses

MTES/SAH Draft v1.5
Currently exact prediction of springing (and whipping) responses of ship is not possible and the current hydroelastic
analysis programs give a good indication of whipping and springing, but not an accurate value for the magnitude of the
springing response primarily due to the uncertainties in the structural damping of each of the key vibrational modes.
Hence it may be necessary to calibrate the hydroelastic ship motion program against some known reference ship.
Alternatively a sensitivity study on the magnitude of the applied structural damping factors can be adopted. So see how
sensitive the results of the fatigue assessment are to the applied structural damping factors. If they are not sensitive, then
the exact value of structural damping is not that important for the fatigue assessment.
If a calibration process is necessary, then a springing analysis for an existing ship where full scale measurements
including springing information have been recorded should be undertaken. Ideally the existing ship should be as similar
as possible to the new design being assessed.
The calibration process can consider the following factors:

Adjustment of the structural damping to achieve the correct magnitude of springing response.
Adjustment of the hull girder structural properties to refine the vibration frequencies, but this is likely to be a
function of incomplete structural modelling
A hydroelastic analysis has to be performed for the reference ship. Hence knowledge of the loading condition, drafts,
mass distribution, speed, heading and environmental conditions (sea state including spectral energy distribution and
directional content) together with hull girder stresses is necessary for several time periods when springing and/or
whipping was present. The hydroelastic analysis should be performed for several of these time periods and the
hydroelastic analysis is to be adjusted to achieve a calibrated springing response. The resultant adjustments can then be
applied to the new ship design.
Various figures in this document illustrate comparisons between full scale measurements and hydroelastic analyses.

MTES/SAH Draft v1.5
Figure 18 Flowchart of the assessment of fatigue using RAOs including springing

responses

MTES/SAH Draft v1.5
7 Alternative Non-linear Springing Fatigue Assessment Methods
7.1 Springing responses using time domain analysis
Alternatively the springing responses can be predicted using a time domain hydroelasticity program. With this method,
it is possible to include non-linearities of the hull form which may be more appropriate for highly flared hull forms.
The time domain method allows the assessment of sub-harmonics or super-harmonics of the springing response which
are not available from most linear frequency domain codes. If an elastic body is excited at twice or three times its
natural frequency, then this will result in oscillation of the body at it natural frequency as well as at the exciting
frequency, this is a sub-harmonic response. The magnitude of oscillation is much weaker than when excited at its
natural frequency, but if there is sufficient energy at the sub-harmonic frequency, then this can cause a significant
oscillation. Similarly if an elastic body is excited at half or one third its natural frequency, then this will result in
oscillation of the body at it natural frequency as well as at the exciting frequency, this is a super-harmonic response. For
the springing response of ships, then we are interested in the super-harmonic frequencies as these have considerable
energy within the sea state energy range.
For example, consider a body with a 2 node vertical bending excitation period of 2 seconds (0.5 Hz, 3.14 rad/s), wave
excitation at periods of 4, 6 and 8 seconds will excite this natural mode. This starts to get important for container ships
at high speed in head seas. For this example, the natural excitation period in the encounter frequency domain is still 2
seconds which at 25 knots in head seas is a wave period is 5.2 seconds. The first three super harmonic periods are still 4,
6 and 8 seconds in the encounter frequency band and these correspond to 8.1, 10.6 and 13.1 seconds in the wave
frequency. As there is considerable wave energy at these periods, this can increase the vertical bending springing
response by over 30%.
A time domain analysis can be used in one of two ways:
1. For analysis of the total response in an irregular wave system corresponding to a sea state.
The time domain hydroelastic method is applicable for the springing fatigue reduction factor method in 6.3.
For the time domain hydroelastic assessment in a sea state, the assumptions discussed in 4.8 should be applied. The
results of this analysis will be a time record of the vertical bending moment or similar. For fatigue purposes, this can be
treated as a time trace and apply the rainflow method is applied to calculate the fatigue index. The rainflow method is
better if there are strong non-linear hull form issues. The alternative is to use a multi peak spectral approach, in this case
the time domain trace needs to be converted to the frequency domain using Fast Fourier Transform (FFT) techniques to
extract the spectral response curve.
FFT methods for limited amounts of data give very spiky power density functions (spectral response curves) due to the
spectral leak and suitable methods, e.g., suitable FFT window functions, should be applied to smooth the spectral
response curves.
2. For analysis of the wave and springing response due to a regular wave
The time domain hydroelastic method can also be applied to regular waves. This approach is applicable for the critical
springing frequency method in 6.2.3, and is discussed further in 7.2.
The normal ship speed and head seas wave condition should be used. For each regular wave a suitable wave height
needs to be selected, this value should represent the typical amount of wave energy of a typical sea state that is critical
for fatigue.

MTES/SAH Draft v1.5
The resulting time domain load response to the regular wave including springing is illustrated in Figure 19 with its FFT
transformation. This example shows the primary wave load response to the regular wave at 1 and two springing
frequencies at s1 and s2, hence wave excitation of the ship at this frequency also excites springing modes and this
wave frequency would be considered a critical wave frequency, see 7.2.
Figure 19 Calculations of time domain springing responses for a regular wave
This method can also be used for the springing RAO method in 6.2.2, but this is really practical applicable only when
vertical bending is the critical springing issue. To derive the complete load RAO curve including springing calculations,
frequencies covering the whole range need to be performed.
If non-linear hull form issues are included, then the responses may need to be linearised in the wave frequency range to
derive the RAO curve.
7.2 Assessment of the springing response to critical wave frequencies
7.2.1 Overview of springing fatigue method using the critical wave frequency method
As noted in 6.2.3, the springing response is dominated by certain exciting frequencies and the total springing response
can be calculated by adding the contribution from each of these frequencies. The contribution to springing from other
wave frequencies is negligible and will not contribute any significant energy to the springing response
The critical exciting wave frequencies are defined as follows:

The wave encounter frequency that matches the natural vibration period.
The wave encounter frequency that matches multiples of the natural vibration period.
The method to determine the critical exciting wave frequencies is given in 7.2.2. The fatigue assessment process is
illustrated in Figure 20.
A time domain linear hydroelastic analysis is undertaken at each of the critical wave frequencies in order to derive the
response of the global loads, see 7.1.

MTES/SAH Draft v1.5
For any sea state, the total vertical bending springing only response Sp is obtained as follows:
N K
S p S RAOsp2 ,n ek sp
n 1 k 1
Where
S() wave energy for the sea state being considered
RAO sp , n ek unit springing response of natural vibration mode n
ek th
k multiple of the natural vibration period in the encounter frequency domain taking account of speed and
heading
sp springing bandwidth factor. Key additional steps are required to evaluate the bandwidth of the natural
vibration period and to evaluate how sensitive the springing response is to small differences in the
excitation frequency. So for example: the critical wave frequency may be 1.0 rad/s, but frequencies of
1.05 rad/s and 0.95 rad/s also excite say 10% of the response at the frequency 1.0 rad/s. This factor is
dependent on the interval between frequencies used for the analysis. This is illustrated in Figure 13 by the
breadth of the springing responses.
The wave energy part of the vertical bending response can be calculated using a conventional rigid body hydrodynamic
code and the springing response is added to this to generate the total vertical bending spectral response curve.
From this a fatigue calculation can be done by using a multi peak short term spectral fatigue calculation or by
transforming the frequency domain results into the time domain and using rainflow counting methods.
A calibration method similar to that discussed in 6.9 will need to be applied.

MTES/SAH Draft v1.5
Figure 20 Flowchart of the springing response to the critical wave

frequencies and the application to the fatigue assessment
7.2.2 Critical wave frequencies for excitation of hull girder vibration modes
All structural systems have many natural vibration modes and any cyclic load (and harmonics of this cyclic load) that
are nearly in resonance with a natural vibration mode will result in these natural vibration modes being excited if the
magnitude of the cyclic load is sufficient.

MTES/SAH Draft v1.5
For a ship structure, excitation of a hull girder natural vibration mode can occur when waves have an encounter
frequency (e) close to a natural frequency (s) of the ships hull girder.
The major component of the springing response occurs when the wave encounter frequency equals or is close to the first
(2 node) hull girder vertical bending hull girder natural frequency of the ship. For ships with large deck openings, then
the coupled 1-node torsional and 2-node horizontal bending vibration mode may have a lower natural frequency and
hence springing of this mode may be a major response.
Major springing excitation frequency, assuming the 2 node vertical bending mode is given by:
s 2 e
Where
wave frequency
1 1 4 A e

2A
e encounter frequency
U
e 1 cos
g
U
A cos
g
sn nth natural hull girder vibration frequency, in rad/s
2
Tsn nth natural hull girder vibration period, in seconds

sn
n vibration mode number, eg 2, 3, 4 etc.
s2: 2 node vertical hull girder natural frequency
s3: 3 node vertical hull girder natural frequency
k multiple of natural hull girder vibration period. k values of 1 to 4 are usually sufficient.
U ship speed (m/s)
ship heading angle, 180o for the head sea.
The same equation applies for other vibration modes such as the 3 node vertical and also the torsional and horizontal
vibration modes.
In addition springing response is also excited by waves which have an encounter period which is a multiple of the hull
girder natural period; however this excitation is a lot less than when the encounter frequency equals the hull girder
natural frequency. These secondary encounter frequencies are given by:
sn k e
Springing can also be excited by nonlinearities of incident wave system. Non-linear waves can be described in terms of
higher order wave cyclic frequencies and the 2nd and 3rd or other higher order wave components may excite springing
responses but these are effects are much smaller and may be ignored.
For head seas and bow quartering sea, the 2 node vertical bending mode is the most critical vibration mode, the 3 node
vertical bending mode is also possible, but this is not so important.
For oblique seas, the first combined 2 node horizontal and 1 node torsional vibration mode of ships with large deck
openings is the most important. However the 1 node torsional mode and 1st (2 node) horizontal bending mode may also
be significant. More complex modes have higher natural frequencies and higher damping and are therefore not so
important.

MTES/SAH Draft v1.5
An example of the relationship between the wave frequencies and vertical bending vibration modes of the hull girder is
given in Table 3.1. The wet mode natural frequencies have been taken from a large containership analysis. The table
gives the critical wave frequencies for the excitation of the first 3 vertical bending vibration modes. These have been
derived by the following relationship:
4 A sn
1 1
k
rad/s
2A
This table shows that at 20 knots in head seas;

Waves with a period of 5.0 seconds will excite the 2 node vertical bending.
Waves of periods 7.8, 10.4 and 12.9 seconds may excite the 2 node vertical bending mode.
Some excitation of the 3 node vertical bending mode may also occur due to the 3rd and 4th multiples of the 3
node 0.9 second vibration period.
Table 6 Example of the critical wave frequencies for exciting different vibration modes.
Wave frequency (rad/s) that matches the kth multiple of the
Vibration frequency wsn
Vibration mode vibration period at a speed of 20 knots and heading of 180 deg
Hz rad/s k=1 k=2 k=3 k=4
2 node (n=2) 0.47 2.953 1.267 0.802 0.603 0.488
3 node (n=3) 1.17 7.351 2.212 1.454 1.124 0.930
4 node (n=4) 2.05 12.881 3.059 2.046 1.601 1.339
Vibration
Wave period (sec) that matches the kth multiple of the vibration
Hz period
period at a speed of 20 knots and heading of 180 deg
Tsn (sec)
2 node 0.47 2.1 5.0 7.8 10.4 12.9
3 node 1.17 0.9 2.8 4.3 5.6 6.8
4 node 2.05 0.5 2.1 3.1 3.9 4.7
Notes
1 The shaded area is the region where there is potentially significant wave energy and hence wave frequencies
that might excite each vibration mode
7.2.3 Example of selection of critical wave frequencies
Using Table 6 as the basis, the following critical wave frequencies and wave heights would be selected for the
assessment:
Table 7 Example of the selection of critical wave frequencies.
Regular wave number Vibration mode and Regular wave frequency Regular wave height
vibration period multiple Hrw
see note 1
RW1 n=2, k=1 1.25 (1.267) See equation below
RW1+d See note 2 1.30 As for RW1

RW1-d See note 2 1.20 As for RW1
Notes
1. Regular wave frequency: Adjusted for the nearest frequency on the assumption of wave frequencies at every

MTES/SAH Draft v1.5
0.025 rad/s.
2. RW1+ d and RW1-d are required to determine the bandwidth correction factor Bsp for the calculation of the
total springing response, see 7.2.1. Note, it may be necessary to consider additional frequencies to determine the
bandwidth correction factor. This is because the natural frequencies obtained by the eigenvalue analysis may be
slightly different to those obtained by the time domain simulations due to the different analysis methods (and
added mass calculations) involved.
The wave height for the regular wave is to be taken as:

H rw 8 S ( ) d
Where
S() spectral energy for a sea state that will be dominant for the springing fatigue assessment
d spacing of wave frequencies assumed in the hydrodynamic analysis.
7.2.4 Calculation of the total rigid body and springing response for the fatigue analysis.
The total value of the wave energy excited response and the springing excited response is calculated by adding the total
springing response due to the sea state energy at the critical wave frequencies and the rigid body response in the sea
state. This is illustrated in Figure 21.
The response energy of the rigid body in a sea state is:
S ( ) R ( ) d
2
RS w ( )
The total springing response energy for each vibrational modal shape in a sea state is:
J
RS spring ( n ) S ( ) R sn2 , j ( ) sp
j 1
This springing response energy can be considered as a single value of energy at the specified springing frequency for
the fatigue calculation.
Total response energy is:

N
RS total ( ) RS w ( ) RS
n 1
spring ( n )
where
R() Rigid body RAO for the appropriate speed and heading
Rsn,j() Springing response for the jth critical wave frequency for vibration mode n
S() Wave energy of the sea state
N is the number of mode shapes being considered
J Number of critical waves being considered for each mode shape
sp Springing bandwidth factor

MTES/SAH Draft v1.5
Response
energy
spectrum
S2 S3
Figure 21 Example of the total response energy spectrum including springing
(figure to be improved)

MTES/SAH Draft v1.5
Lloyds Register EMEA Lloyds Register ASIA Lloyds Register Americas, Inc
71 Fenchurch Street 22nd Floor, Dah Sing 1330 Enclave Parkway, Suite 200
London EC3M 4BS, UK Financial Centre Houston
108 Gloucester Road Texas 77077
T +44 (0)20 7709 9166 Wanchai, Hong Kong USA
F +44 (0)20 7488 4796 SAR of PR China
E emea@lr.org T +1 (1)281 675 3100
T +852 2287 9333 F +1 (1)281 675 3138
www.lr.org F +852 2845 2616 E americas-marine@lr.org
E asia@lr.org
www.lr.org
www.lr.org
Lloyds Register is a trading name of Lloyds Register Group Limited and its subsidiaries. For further details please see http://www.lr.org/entities
Lloyd's Register Group Limited, its affiliates and subsidiaries and their respective officers, employees or agents are, individually and collectively, referred to
in this clause as Lloyd's Register. Lloyd's Register assumes no responsibility and shall not be liable to any person for any loss, damage or expense caused
by reliance on the information or advice in this document or howsoever provided, unless that person has signed a contract with the relevant Lloyd's
Register entity for the provision of this information or advice and in that case any responsibility or liability is exclusively on the terms and conditions set
out in that contract.
Lloyds Register, 2014
Working together
for a safer world

213-43556 LR Guidance Notes On Springing and Whipping v1 5 Draft.5 Draft

Diunggah oleh

Informasi Dokumen

Judul Asli

Hak Cipta

Format Tersedia

Bagikan dokumen Ini

Bagikan atau Tanam Dokumen

Opsi Berbagi

Apakah menurut Anda dokumen ini bermanfaat?

Apakah konten ini tidak pantas?

Hak Cipta:

Format Tersedia

213-43556 LR Guidance Notes On Springing and Whipping v1 5 Draft.5 Draft

Diunggah oleh

Hak Cipta:

Format Tersedia

ShipRight

Design and Construction

Structural Design Assessment

Guidance Notes on the Assessment of Global

Lloyds Register 1 June 2014

4.3 Derivation of the EDSs 23

Lloyds Register 2 June 2014

Lloyds Register 3 June 2014

1.1 Guidance Document

1.2.2 Assessment of non-linear ship motion wave loads

Lloyds Register 4 June 2014

b) RABF > 0,2 or

1.2.3 Assessment of whipping

1.2.4 Assessment of springing responses including the effect on fatigue

Lloyds Register 5 June 2014

1.3 Non-linear effects

1.4 Whipping Phenomena

Lloyds Register 6 June 2014

Figure 2 Typical whipping effects of VBM in the full scale measurement.

Lloyds Register 7 June 2014

1.6 Hull girder natural vibration frequencies

1.7 Assessment Procedure

1.7.1 Assessment of the long term design linear wave loads

1.7.2 Assessment of the non-linear ship motion wave loads

Lloyds Register 8 June 2014

Linear Design Loads

Ship Motion and

Environmental Linear Design

Nonlinear Design Loads Nonlinear Design Loads

Nonlinear Sag/ Whipping Sag/

1.7.3 Assessment of the hull girder loads including whipping loads

1.7.4 Assessment of fatigue including hull girder springing loads

1.8 Documentation and plans to be submitted

Lloyds Register 9 June 2014

Lloyds Register 10 June 2014

2 Assessment of the long term linear design wave loads

2.2 Loading conditions

2.2.1 Loading conditions for the whipping and non-linear assessments

2.2.2 Loading conditions for the springing assessment

Lloyds Register 11 June 2014

2.3 Hydrodynamic model

2.4 Verification of hydrodynamic models

Lloyds Register 12 June 2014

2.5 Roll damping coefficient

2.6 Ship operational conditions

Lloyds Register 13 June 2014

2.7 Ship motion and load responses in frequency domain

Table 3 Ship motion analysis parameters

Loading conditions As specified in 2.2

Speed As specified in 2.6

2.8 Wave environment

Lloyds Register 14 June 2014

2.9 Linear design wave bending moment

2.10 Long term calculation

height and averaged zero-crossing wave period ( H s , Tz ) .

Lloyds Register 15 June 2014

3 Assessment of non-linear ship motion wave loads

Linear nonlinear whip

Figure 4 Vertical bending moment amidships, = 0.91Lpp

3.2 Determination of the Equivalent Design Wave (EDW)

Lloyds Register 16 June 2014

Please note the recommended method is to use the EDS approach.

3.3 Determination of the Equivalent Design Sea State (EDS)