Optimization of Bow Shape for Large,

Slow Ships

Daniel Edward Nordås

Marine Technology
Submission date: June 2012
Supervisor: Sverre Steen, IMT

Norwegian University of Science and Technology

Department of Marine Technology
PROBLEM DEFINITION
MASTER THESIS IN MARINE TECHNOLOGY

SPRING 2012

FOR

Daniel Nordås

Optimization of bow shape of large slow ship

With respect to green shipping, the main challenge for the hydrodynamics community is to help
reducing the consumption of fossil fuels. Traditionally, ships have been optimized for minimum fuel
consumption in calm water. For large slow ships like VLCCs, this has led to very blunt bow shapes.
Such bow shapes have high added resistance due to waves. Thus, one might think that the optimum
bow shape, when realistic wave conditions are taken into account, should be more slender than the
current shapes. Furthermore, the operational area of the ship (the route it sails) could influence what is
the optimum bow shape. The objective of the thesis is therefore to explore optimum bow shapes for
large slow ships (like large tankers and bulkers), taking typical wave conditions into account.

To reach the objective, the candidate is recommended to do the following:

 Review previous scientific work on design of hull shape of large slow ships like tankers and
bulkers.
 Establish a method to predict the resistance of such hull forms in calm water and in waves (or
added resistance due to waves). Verify that the method gives reliable results by comparing
with benchmark data.
 Identify a number of different routes, typical for large tankers and bulkers, representing
different operational conditions with respect to waves. Use completely calm water as a
reference route.
 Optimize the bow shape for minimum resistance (or power consumption) on the different
routes. Discuss how the optimum bow shape is influence by the choice of route.
 Several innovative bow shapes to reduce added resistance in waves, which mainly involves re-
designing the above-water area, have been proposed, like “ax-bow”, “beak-bow” and “X-
bow”. Give a review of these bow shapes, and discuss their working principles and which of
them that look promising for your ships.
 If time allows, evaluate the efficiency of one such bow design.

In the thesis the candidate shall present his personal contribution to the resolution of problem
within the scope of the thesis work.
Theories and conclusions should be based on mathematical derivations and/or logic reasoning
identifying the various steps in the deduction.

The candidate should utilize the existing possibilities for obtaining relevant literature.

The thesis should be organized in a rational manner to give a clear exposition of results,
assessments, and conclusions. The text should be brief and to the point, with a clear language.
Telegraphic language should be avoided.

The thesis shall contain the following elements: A text defining the scope, preface, list of
contents, summary, main body of thesis, conclusions with recommendations for further work,
list of symbols and acronyms, reference and (optional) appendices. All figures, tables and
equations shall be numerated.

The supervisor may require that the candidate, in an early stage of the work, present a written
plan for the completion of the work. The plan should include a budget for the use of computer
and laboratory resources that will be charged to the department. Overruns shall be reported to
the supervisor.

The original contribution of the candidate and material taken from other sources shall be
clearly defined. Work from other sources shall be properly referenced using an acknowledged
referencing system.

The thesis shall be submitted in two copies:

- Signed by the candidate

- The text defining the scope included
- In bound volume(s)
- Drawings and/or computer prints that cannot be bound should be organized in a
separate folder.
- The bound volume shall be accompanied by a CD or DVD containing the written
thesis in Word or PDF format. In case computer programs have been made as
part of the thesis work, the source code shall be included. In case of experimental
work, the experimental results shall be included in a suitable electronic format.

Supervisor : Professor Sverre Steen

Advisor : Olav Rognebakke (DNV)
Start : 16.01.2012
Deadline : 10.06.2012

Trondheim, 16.01.2012

Sverre Steen
Supervisor
PREFACE
This thesis is submitted to the Norwegian University of Science and Technology (NTNU) as a
requirement for the degree of Master of Science. The workload is equivalent to 30 credits, or one
semester, at NTNU.

This Master thesis has been performed at the Department of Marine Technology, NTNU,
Trondheim, with Professor Sverre Steen as supervisor, and Olav Rognebakke, DNV, as advisor.

During the work with this thesis several software packages like Friendship/Shipflow, Maxsurf
Pro, ShipX and Rhinoceros that were more or less new to me have been utilized. Especially the
work with Friendship/Shipflow has been very time consuming. Due to lack of experience with
Friendship/Shipflow everything from generating geometry offsets to the setup for generating
calm water wave resistance has been an iterative process with a lot error-searching. Poor results
have resulted in a lot of testing with little success.

The last objective in the problem definition asks for an evaluation of the efficiency of innovative
bow designs by computational fluid dynamics. Due to the unforeseen time demanding work with
the Shipflow this objective has not been answered and has been stated as further work.

Many different sources of information have been utilized throughout the work of this thesis. I
would like to thank Jaeouk Sun at DNV for sharing his knowledge on hull design and his help with
Friendship/Shipflow as well as discussions on the results and all the people at Ship
Hydrodynamics and Stability that has motivated and willingly helped with issues throughout the
work with this thesis. I would also like to thank Olav Rognebakke for sharing his insight to the
investigated problem and introducing me to the correct persons to ask. Lastly, I would thank Prof.
Sverre Steen for sharing his knowledge and opinions on the subject and being of great assistance
in structuring and discussing different topics in this thesis.

Trondheim, 08.06.2012

Daniel E. Nordås
ABSTRACT
Traditionally ships have been optimized for minimizing the fuel consumption in calm water. For
slow, large ships like tankers and bulk carriers this has resulted in very blunt bows with high
added resistance due to waves. The objective of this thesis has been to investigate if the optimal
bow shape, when realistic wave conditions are taken into account, should be more slender than
the current blunt bows. The added resistance is also highly dependent on the actual wave
conditions the vessel experiences. Thus a question has been if the optimal bow changes with the
operational area, or route, of the vessel.

Five designs have been investigated representing a range of waterlines from blunt to sharp. They
are based on the MOERI Tanker KVLCC2. KVLCC2A is the original design of the MOERI Tanker
with no flare. KVLCC2B has the same water line curve as KVLCC2A, but with straight sides and
small bilge radius in the bow. KVLCC2C has a more slender bow by moving volume from the
shoulders to above the bulb. KVLCC2D is a blunter design than KVLCC2A and KVLCC2E has been
elongated by 8 m compared to KVLCC2C to get a more slender bow.

Four routes have been chosen to represent trades and ocean areas. The routes are; Arabian Gulf
(AG) to the Gulf of Mexico (GM), AG to Japan, Brazil to China and Norway to the East Coast of US.

Calm water resistance has been calculated and verified against experimental data. The wave
resistance was calculated numerically using Shipflow. These calculations were not satisfying and
should be taken a closer look at. Modification of the results had to be done.

The results show that KVLCC2A, KVLCC2C and KVLCC2E have very similar calm water resistance.
They have slightly lower values than KVLCC2D. KVLCC2B has the greatest calm water resistance.

The added resistance was calculated by ShipX. The sharper bow designs have significantly lower
resistance in the diffraction regime, as intended. KVLCC2E has slightly a slightly lower added
resistance coefficient in the short wave regime than KVLCC2C.

The speed-loss calculations were performed by combining wave statistics for the routes, calm
water resistance, added resistance and engine and propulsion characteristics in ShipX. The result
is an attainable speed at a given power input, 27 000 kW.

The results show that KVLCC2C and KVLCC2E have the lowest speed-loss. The attainable speed is
highest for KVLCC2C and it can thus be concluded that a sharper design is more optimal when
realistic wave conditions are taken into account.

The relative speed loss on different routes between KVLCC2C and KVLCC2A shows that the speed
loss of KVLCC2C is 14.2% lower for the AG to GM, 13.8%, 16.2 % and14.9 % for respectively AG
to Chiba, Mongstad to East coast of US and Brazil to China. Thus, a small difference can be seen,
but not enough to change the best design in this case.

A review of innovative bow shapes dealing with added resistance was performed and an
evaluation based on working principles and applicability to a large, slow vessel was discussed.
The designs reviewed were X-bow (Ulstein Design), a new bow from STX OSV and Beak-bow, Ax-
bow and LEADGE-bow designed in Japan especially for larger ships.

The two first bows are designed primarily with offshore service vessels in mind and focus more
on the long waves. The LEADGE-bow, which is based more or less on the same principles as
KVLCC2C, shows that a simple sharpening of the bow is an easy and effective measure. This
seems like the most promising bow for large, slow ships of those evaluated.

iii
SAMMENDRAG
Historisk har skip blitt optimalisert for å minimere drivstofforbruket i stille vann. For store,
saktegående skip, som tankere og bulkskip, har dette resultert i veldig butte baugformer med høy
tilleggsmostand i bølger. Formålet med denne oppgaven har vært å undersøke om slankere
baugformer, når realistiske bølgetilstander er tatt i betraktning, kan være mer optimalt enn de
tradisjonelle butte baugene. Tilleggsmotstand er også svært avhenging av hvor store de faktiske
bølgetilstandene er, og et spørsmål har vært om den optimale baugformen kan være avhengig av
operasjonsområdet (ruten) et skip trafikkerer.

Fem design, som representerer variasjon fra butt til slank baug, har blitt undersøkt. De er basert
på MOERI tankeren KVLCC2. KVLCC2A er originaldesignet av MOERI tankeren, men med rette
sider over vann. KVLCC2B har samme form i vannlinjen som KVLCC2A, men med rette sider
under og over vann. KVLCC2C har en slankere baug, formet ved å flytte volum fra fremre skulder
til tomrommet over bulben på KVLCC2A. KVLCC2D har en enda buttere baug enn KVLCC2A.
KVLCC2E har blitt forlenget med 8 meter i forhold til KVLCC2C for å få en slankere baug.

Fire har blitt valgt ut for å representerer handelsruter og bølgetilstander. Rutene er den Arabiske
Gulfen (AG) til Mexico Gulfen (MG), AG til Japan, Brasil til Kina og (Norge) til Østkysten av USA .

Stillevannsmotstanden har blitt beregnet og verifisert mot eksperimentelle data.

Bølgemotstanden ble kalkulert numerisk ved hjelp av Shipflow. Disse beregningene ga ikke
tilfredsstillende resultater og burde bli sett nøyere på. Modifikasjon av resultatene har blitt gjort.

Resultatene viser at KVLCC2A, KVLCC2C og KVLCC2E har veldig lik stillevannsmotstand og

resultatene ligger litt under motstanden til KVLCC2D. KVLCC2B har størst stillevannsmotstand.

Tilleggsmotstanden ble beregnet ved hjelp av ShipX. De slankeste baugformene har signifikant
reduksjon av tilleggssmotstanden i korte bølger, som ønsket. KVLCC2E har litt lavere
tilleggsmotstand i korte bølger enn KVLCC2C.

Beregning av fartstapet ble beregnet ved å kombinere bølgestatistikk, stillevanns- og tilleggs-

motstand og maskineri og propulsjons egenskaper i ShipX. Resultatet er presenter i form av et
fartstap og en oppnåelig hastighet for rutene ved en gitt effekt på maskineriet, 27 000 kW.

Resultatene viser at KVLCC2C og KVLCC2D har de laveste fartstapene. Den oppnåelige

hastigheten er høyest for KVLCC2C og det kan dermed konkluderes med at et noe slankere design
er mer optimalt når realistiske bølgedata er tatt i betraktning.

Det relative fartstapet på forskjellige ruter mellom KVLCC2A og KVLCC2C viser at KVLCC2C har
14.2 % mindre fartstap på ruten fra AG til MG, 13.8 %, 16.2% og 14.9% for rutene fra henholdsvis
AG til Japan, Norge til østkysten av USA og Brasil til Kina. En liten forskjell kan altså ses på
forskjellige ruter, men dette påvirker ikke resultatet i denne oppgaven.

En betraktning av nye innovative baugformer, med hensikt å redusere totalmotstanden i bølger,

har også blitt gjennomført. I tillegg har anvendbarhet på store, saktegående skip blitt diskutert.
Designene som har blitt undersøkt er X-bow (Ulstein Design), STX sin nye baugform og Beak-bow,
Ax-bow og LEADGE-bow designet i Japan spesielt med tanke på større skip.

De to første baugformene er hovedsakelig designet med tanke på offshore service skip og fokuset
ligger mer på tilleggsmotstanden i lange bølger siden disse vanligvis er betraktelig mindre enn
store tankere og bulkskip. LEADGE-bow, som er baser på mer eller mindre de samme
prinsippene som KVLCC2C, viser at et enkelt tiltak som en slankere baug er effektivt. Dette virker
også som den mest lovende baugen for store skip av de som ble evaluert.

iv
CONTENTS
Preface ___________________________________________________________________________________ iii
Abstract __________________________________________________________________________________ iii
Sammendrag ____________________________________________________________________________ iv
1 Background and Motivation ______________________________________________________ 1
1.1 Hypothesis ____________________________________________________________________________1
1.2 Scope Of Work________________________________________________________________________3
1.3 Added Resistance Due to Waves __________________________________________________4
1.4 Calculation of Added Resistance in Waves ______________________________________5
1.5 Discussion of Methods ______________________________________________________________9
2 Methods_____________________________________________________________________________ 10
2.1 Calm Water Resistance ___________________________________________________________ 10
2.2 Added Resistance in Waves ______________________________________________________ 13
2.3 Added Resistance due to wind ___________________________________________________ 13
2.4 Route Simulation __________________________________________________________________ 13
2.5 Designing the Models _____________________________________________________________ 14
3 Changing Bow Design ____________________________________________________________ 15
3.1 Resistance Considerations _______________________________________________________ 15
3.2 Length and Displacement Considerations _____________________________________ 15
3.3 The Designs _________________________________________________________________________ 18
4 Routes and Wave Statistics ______________________________________________________ 26
4.1 Routes _______________________________________________________________________________ 26
4.2 Wave Statistics _____________________________________________________________________ 27
5 Analysis and Results ______________________________________________________________ 35
5.1 Verification _________________________________________________________________________ 35
5.2 Results _______________________________________________________________________________ 41
6 Conclusion__________________________________________________________________________ 51
7 Innovative Bow Designs _________________________________________________________ 53
7.1 X-Bow ________________________________________________________________________________ 53
7.2 STX Design __________________________________________________________________________ 55
7.3 Ax-Bow ______________________________________________________________________________ 56
7.4 Beak Bow ___________________________________________________________________________ 58

v
 7.5 Leadge Bow _________________________________________________________________________ 59
7.6 Evaluation of Designs Implemented on KVLCC2 ______________________________ 60
8 Further Work ______________________________________________________________________ 62

vi
LIST OF FIGURES
Figure ‎1.1 - Wave reflection off blunt bow _____________________________________________________________2
Figure ‎1.2 - Wave reflection off sharp bow _____________________________________________________________2
Figure ‎1.3 - Added resistance of KVLCC2 model in waves, (Guo, 2011). ________________________________4
Figure ‎1.4 - Components of added resistance and their conceptual relative significance at different
wavelengths (Hirota et al., 2005) _______________________________________________________________________5
Figure ‎2.1 - ITTC-curve for friction ___________________________________________________________________ 11
Figure ‎3.1 - Schemtaically options of changing the bow shape. Black line is original bow shape, blue
line is decreased displacement and red line is lengthened ship. ______________________________________ 16
Figure ‎3.2 - Combination of the two designs. The volume moved from 1 and 2 is equal to volume 3 17
Figure ‎3.3 - The hatched area can be used to move volume forward and making the bow sharper_ 17
Figure ‎3.4 - KVLCC2A body blan ______________________________________________________________________ 18
Figure ‎3.5 - KVLCC2B body plan ______________________________________________________________________ 19
Figure ‎3.6 - KVLCC2C body plan ______________________________________________________________________ 20
Figure ‎3.7 - KVLCC2D body plan ______________________________________________________________________ 21
Figure ‎3.8 - KVLCC2E body plan ______________________________________________________________________ 22
Figure ‎3.9 - Waterlines of the designs ________________________________________________________________ 23
Figure ‎3.10 - Sectional area curve of KVLCC2A ______________________________________________________ 23
Figure ‎3.11 - Sectional area curve of the bow area of all designs ____________________________________ 24
Figure ‎3.12 - Relative performance between designs regarding frictional resistance and added
resistance _____________________________________________________________________________________________ 24
Figure ‎4.1 - Map of predefined areas and routes. Red lines is VLCC routes and yellow is the iron ore
bulk route. The green is from Mongstad to New York ________________________________________________ 28
Figure ‎4.2 - Weighted scatter diagram from Ras Tanura to Chiba __________________________________ 30
Figure ‎4.3 - The cumulative distribution function in percent – Ras Tanura – Chiba ________________ 31
Figure ‎4.4 – Weighted scatter diagram from Ras Tanura to LOOP __________________________________ 32
Figure ‎4.5 – The cumulative distribution function in percent – Ras Tanura to LOOP _______________ 32
Figure ‎4.6 – Weighted scatter diagram from Mongstad to New York _______________________________ 33
Figure ‎4.7 - The cumulative distribution function in percent – Mongstad to New York _____________ 33
Figure ‎4.8 - Weighted scatter diagram from Tubarão to Qingdao ___________________________________ 34
Figure ‎4.9 - The cumulative distribution function in percent –Tubarão to Qingdao ________________ 34
Figure ‎5.1 - Rw from numerical calculations and Rr from experiments ______________________________ 37
Figure ‎5.2 - RW from wave cuts in Shipflow ___________________________________________________________ 37
Figure ‎5.3 - RW from integrated pressure in Shipflow ________________________________________________ 38
Figure ‎5.4 – Comparison of the model test residual resistance and modified residual resistance ___ 39
Figure ‎5.5 - Experimental and numerical results for added resistance for KVLCC2A (Guo, 2011) __ 40
Figure ‎5.6 - Added resistance coefficient for KVLCC2A from ShipX at FN=0.142 _____________________ 41
Figure ‎5.7 - Modified residual resistance for a small range of Froude numbers _____________________ 41
Figure ‎5.8 - Viscous resistance for the designs _______________________________________________________ 42
Figure ‎5.9 – Total calm water resistance _____________________________________________________________ 43
Figure ‎5.10 - Total calm water resistance plotted against knots ____________________________________ 43
Figure ‎5.11 - RAO in added resistance for KVLCC2A at different headings, FN=0.142 _______________ 44
Figure ‎5.12 - Added resistance RAO at head seas at FN = 0.142 ______________________________________ 44
Figure ‎5.13 - Comparison of added resistance calculations in ShipX for KVLCC2B models, FN=0.14245

vii
Figure ‎5.14 - Engine size of comparison ships from Sea-Web. _______________________________________ 46
Figure ‎5.15 - KVLCC2A: Speed-loss over headings for each route ____________________________________ 47
Figure ‎5.16 - KVLCC2B: Speed-loss over headings for each route ____________________________________ 47
Figure ‎5.17 - KVLCC2C: Speed-loss over headings for each route ____________________________________ 47
Figure ‎5.18 - KVLCC2D: Speed-loss over headings for each route ____________________________________ 48
Figure ‎5.19 - KVLCC2E: Speed-loss over heading for each route _____________________________________ 48
Figure ‎5.20 – Speed-loss for each design on the route from the Arabian Gulf to LOOP ______________ 48
Figure ‎5.21 – Speed-loss averaged over headings grouped by route _________________________________ 49
Figure ‎5.22 - Speed-loss averaged over headings grouped by design ________________________________ 49
Figure ‎5.23 - Attainable speed in waves averaged over headings grouped by route ________________ 49
Figure ‎5.24 - Attainable speed in waves averaged over headings grouped by design _______________ 50
Figure ‎7.1 - The X Bow concept _______________________________________________________________________ 53
Figure ‎7.2 - Waterline of X-Bow, (Kvamsvåg, 2006). _________________________________________________ 54
Figure ‎7.3 - STX OSV - PSV 08 (http://www.stxosv.com/newsandmedia/Pages/default.aspx) _____ 55
Figure ‎7.4 - Profile view of the STX bow, (Tvete and Borgen, 2012). _________________________________ 56
Figure ‎7.5 - The Ax-bow concept compared to the ordinary bow and LEADGE-bow (ITTC, 2008). _ 57
Figure ‎7.6 - Added resistance results from model tests with Ax-bow ________________________________ 58
Figure ‎7.7 - Beak bow profile and waterplane view, (Hirota et al., 2005). ___________________________ 59
Figure ‎7.8 - Full scale measurements of Ax bow fitted on a cape-size bulk carrier, (Hirota et al.,
2005) __________________________________________________________________________________________________ 60

viii
LIST OF TABLES
Table ‎1.1 - Main dimensions of the original KVLCC2 design ___________________________________________3
Table ‎3.1 - Main dimensions KVLCC2A _______________________________________________________________ 18
Table ‎3.2 - Main dimensions KVLCC2B _______________________________________________________________ 19
Table ‎3.3 - Main Dimensions KVLCC2C _______________________________________________________________ 20
Table ‎3.4 - Main dimensions KVLCC2D _______________________________________________________________ 21
Table ‎3.5 - Main dimensions KVLCC2E _______________________________________________________________ 22
Table ‎4.1 - Area and ratios used as input to RouteSim _______________________________________________ 28
Table ‎4.2 - Total distances (source: http://www.sea-distances.com/) ______________________________ 28
Table ‎4.3 - Percentage of sea states within the limitations of Faltinsen's formula for added
resistance in short waves _____________________________________________________________________________ 30
Table ‎5.1 - Main dimensions of ship and model used in MARINTEK experiments____________________ 35
Table ‎5.2 - Results from MARINTEK experiments, form factor k = 0.2021 ___________________________ 35
Table ‎5.3 - Results from MARINTEK experiments with form factor from numerical calculations
(Steen et al., 2010) and empirical form factor by Holtrop. ___________________________________________ 36
Table ‎5.4 - Viscous resistance for KVLCC2A, form factor = 0.3435 __________________________________ 39
Table ‎5.5 - Form factors calculated by Holtrop's method ____________________________________________ 40
Table ‎5.6 - Percantage change in viscous resistance calculated in newton at FN=0.142, with
KVLCC2A as base case ________________________________________________________________________________ 42
Table ‎5.7 - Propeller characteristics __________________________________________________________________ 46
Table ‎5.8 - Propulsion settings _______________________________________________________________________ 46
Table ‎5.9 - Attainable speed in calm water based on engine and propulsion configuration presented
in chapter ‎5.2.3.1. _____________________________________________________________________________________ 47
Table ‎5.10 - Percentage speed-loss of attainable speed in calm water ______________________________ 50
Table ‎7.1 - Reduction ratios of speed-loss, (Hirota et al., 2005). _____________________________________ 58

ix
NOMENCLATURE
Symbol SI unit Explanation
Aproj [m2] Projected area of above water hull and superstructure
[deg] Wave propagation direction with respect to the x-axis
Bwl [m] Beam in water line
BZ [N s/m] Heave damping coefficient
[N s/rad] Pitch damping coefficient
[N s/rad] Coupled heave and pitch damping coefficient
CA [-] Correlation allowance
CAA(s,m) [-] Air resistance coefficient for ship or model
CApp(s,m) [-] Appendix resistance coefficient for ship or model
CB [-] Block coefficient
CDB(s,m) [-] Wet transom stern resistance coefficient for ship or model
CF(s,m) [-] Skin friction resistance coefficient for ship or model
CM [-] Midship coefficient
CP [-] Prismatic coefficient
CR [-] Residual resistance coefficient
CStern [-] Coefficient describing the shape of the stern
CT(s,m) [-] Total friction coefficient for ship or model
CVS [-] Viscous resistance coefficient
CX [-] Air resistance coefficient given in ShipX
, , [rad] Heave, pitch and relative motion phase angle
CF [-] Hull roughness allowance
D [m] Moulded depth
Fa [N] Amplitude of heave force
FN [-] Froude number
̅ [N/m] Force per unit length normal to hull
̂ [-] Cumulative distribution function
[N/m] Mean added resistance on a strip
g [m/s2] Gravitational acceleration
H [µm] Hull roughness
HS [m] Significant wave height
k [rad/m] Wave number
k [-] Form factor
kf [-] Form factor due to form effect on friction
kp [-] Form factor due to form effect on pressure
L1 [m] Non-shadow line segment of water line
Lpp [m] Length between perpendiculars
LR [m] Distance from bow to parallel mid ship
Lwl [m] Length in water line
Ma [Nm] Amplitude of pitch moment
n [-] Total number of observations
[-] Number of observations lower or equal to TZ
[1/s] Encounter frequency
3
∇ [m ] Volume displacement
3
ρ [kg/m ] Density
Raw [N] Added resistance
RN [-] Reynolds number

x
RT [N] Total resistance
S [m2] Wetted surface area
SB [m2] Transom stern wetted surface area
s [m] Relative vertical motion amplitude
T [m] Draught
TZ [s] Zero crossing period
[deg] Angle between tangent of water line and centre line
[rad] Pitch amplitude
V [m/s] Speed in m/s or knots
VattCW [m/s] Attainable speed in calm water
xb [m] Longitudinal position of strip
Za [m] Heave amplitude
zx [m] Vertical motion due to pitch
[m] Wave elevation
[m] Wave amplitude

Abbreviation and
Acronyms
AP Aft Perpendicular
CDF Cumulative Distribution Function
CFD Computational Fluid Dynamics
DWT DeadWeight Tonnage
EUT Enhanced Unified Theory
FP Fore Perpendicular
ITTC International Towing Tank Conference
JONSWAP JOint North Sea WAve Project
LCB Longitudinal Centre of Buoyancy
LCG Longitudinal Centre of Gravity
RANS Reynolds Averaged Navier-Stokes
RPM Revolutions per minute
VERES VEssel RESponses (ShipX plug-in)
VLCC Very Large Crude Carrier, 160 000 – 319 999 DWT
VOF Volume Of Fluids

xi
1 BACKGROUND AND MOTIVATION
Historically, merchant ship owners have tended to be conservative and new innovative solutions
have been hard to introduce. Merchant vessel design has focused on low hull resistance and high
propulsion efficiency where the verification process has been model testing in calm water
conditions, (Laursen, 2001).

Added resistance, or involuntary speed-loss, due to incident waves and wind has been included
by adding a sea margin. This sea margin has a typical value between 15%-30% of calm-water
power, based on experience with similar ships (Arribas, 2007). This focus on calm water
resistance has led to hull forms with good performance in calm water, which for large, slow ships
like Very Large Crude Carriers (VLCC) or bulk carriers, have resulted in very blunt bows.

The major resistance component of large, slow ships is the viscous resistance. Wave making
resistance is only a small part of the total resistance component for such ships, due to the low
Froude number, even with blunt bows.

Blunt bows have a larger volume to wetted surface ratio than a sharper bow with equal
displacement. Thus more cargo can be transported per wetted surface area. Also, due to the low
Froude number and hence low ratio of wave resistance, the benefit of decreasing the wave
resistance by sharpening the bow may be cancelled by an increase in viscous resistance or a
decrease in cargo capacity. These two arguments support a blunt bow.

The sheltered calm water conditions the ships are verified for are rare on the typical trade routes
for these types of vessels. Involuntary speed-loss due to wind and waves may slow down the
vessel. Two separate issues arise here. Firstly, the design of the vessel with respect to added
resistance in wind and waves may be poor resulting in significant speed-loss that needs to be
compensated for by increasing the power consumption. Thus, this increases the fuel
consumption i.e. the environmental impact as well as the fuel expenses. Secondly, the speed-loss
may be impossible to compensate for, as the available installed power reserve is too small,
resulting in delays possibly causing economic loss for the ship owner. In addition, if the installed
power reserve is too small, it may cause issues with manoeuvring which may be a critical safety
issue.

In Faltinsen and Svensen (1990) a 198 meter long container vessel was used to investigate the
speed-loss during a round-trip on a North-Atlantic route. They found that on a target speed of 22
knots, the ship had a speed-loss of 1.7 knots westbound and 0.9 knots round-trip voyages due to
involuntary speed-loss. Economically, a speed-loss of 0.5 knots when translated directly into lost
cargo carried represents a total economic loss of ~$2 million over one year (in 1990).

1.1 HYPOTHESIS
Reflection of waves off the bow and radiation of waves due to ship motions are two of the major
effects contributing to involuntary speed-loss due to added resistance in incident waves.
Reflection of waves is dominant when the wavelengths are short compared to the ship length,
while the radiation effect dominates when the wavelengths are in the range of the ship length
(refer to chapter ‎1.3).

The majority of the sea states that a large ship encounters on a voyage are small, and thus is
associated with small wavelengths. Even though the added resistance in magnitude is largest
around resonance in heave and pitch, added resistance due to short waves are important in the
long term (Faltinsen, 2010).

1
One relatively simple measure to reduce the reflection of waves is to sharpen the entrance angle
of the bow, (Guo and Steen, 2010b). Thus the incident waves will be reflected more to the sides
than forward, decreasing the momentum change of the incident waves and thus their impact on
the vessel. This is depicted in Figure ‎1.1 and Figure ‎1.2.

FIGURE ‎1.1 - WAVE REFLECTION OFF BLUNT BOW

FIGURE ‎1.2 - WAVE REFLECTION OFF SHARP BOW

Thus the hypothesis of this thesis is that a possible increase in the calm water resistance due to a
sharper bow, would be compensated, and maybe surpassed by a decrease in the added resistance
due to waves. This may contribute to a vessel with a better total performance; less fuel
consumption, lower emissions and better delivery precision.

However, the result is expected to depend on the trade-off between increased calm water
resistance and decreased added resistance, the time spent in practically calm water and waves
and the degree of severity of the sea states on the routes. A ship trading on a route with

2
practically no waves will be expected to have less benefit from a sharper bow than a ship trading
on routes with severe wave conditions most of the time.

1.2 SCOPE OF WORK

The objective of this thesis is to investigate the performance of bow designs with different
slenderness for a large, slow vessel on a typical route with realistic wave conditions. The aim of
the calm water resistance analysis is to estimate the calm water resistance of each design,
capturing the changes in resistance due to changes in the bow design. The added resistance
analysis is performed to investigate the influence of the bow design on the added resistance in
both short and long waves. Scatter diagrams is established from wave statistics databases to such
that the designs can be evaluated in a realistic wave environment.

The MOERI tanker, KVLCC2, has been chosen as a basis for the designs. This is a VLCC with
dimensions shown in Table ‎1.1. It has become an academic standard for validation of
Computational Fluid Dynamics (CFD), and the original 3D model is provided on the website of
Simman2008 (2008).

TABLE ‎1.1 - MAIN DIMENSIONS OF THE ORIGINAL KVLCC2 DESIGN

Main Particulars KVLCC2 original

Lpp (m) 320.0
Lwl (m) 325.5
Bwl (m) 58.0
D (m) 30.0
T (m) 20.8
Displacement ∇ (m3) 312622.0
S excl. rudder (m2) 27194.0
CB 0.8098
CM 0.9980
LCB (%). fwd+ from
3.4800
amidship
LCB (m) from AP 171.3

The rest of this chapter gives a brief review of the theory of added resistance and a literature
review of work done on added resistance in waves. This review focuses on methods to calculate
the added resistance and ends with a discussion of the most promising methods for this thesis.

Chapter ‎2 establishes and explains the methods and software used to calculate the calm water
resistance, added resistance and the simulation of the performance on typical routes.

Chapter ‎3 discusses the process of developing alternative bow designs and issues with changing
the bow design. The designs used in this thesis is presented and evaluated. In total 5 designs have
been chosen to represent the range from blunt to sharp bows.

Chapter ‎4 establishes typical routes for VLCCs and bulk carriers. The long term wave statistics
will be established with scatter diagrams for each route, representing the probability of
occurrence of sea states on the route.

Chapter ‎5 verifies the methods described in chapter ‎2 and discusses issues regarding the
methods. Calm water resistance, added resistance and finally route simulation results is
presented and discussed.

3
Due to restrictions of methods used in this thesis, the models have straight sides. In chapter ‎7
innovative bow shapes, some with more complex geometry, already developed is presented. The
working principles are discussed and the efficiency is evaluated on the basis of applicability to
large, slow ships like VLCCs.

1.3 ADDED R ESISTANCE DUE TO WAVES

The added resistance is commonly referred to as the difference between the resistance of a ship
in waves and the calm water resistance. Due to the large variations in resistance when the ship is
travelling in waves, the most sensible measure is the average added resistance in a wave
condition, or the involuntary speed-loss caused by the added resistance in waves (refer to
Figure ‎1.3).

FIGURE ‎1.3 - ADDED RESISTANCE OF KVLCC2 MODEL IN WAVES, (GUO, 2011).

The energy dissipated from a ship due to incident waves can, according to classical sea keeping
theories, be split into two main components, (Wilson, 1985).

 The interference between the incident wave system and the radiated waves resulting
from ship motions, where the vertical motions, heave and pitch, are the most important.
This is often referred to as drift force.
 The wave system will experience some wave reflection of the incident waves off the ship.
This component becomes important when the wavelengths are small compared to the
ship length. This component is commonly known as wave diffraction.

When applying analytical consideration, these three components are in principle additive and
can be superimposed. However, in reality they interact and a division in the three components
cannot be strictly observed. In practice, all these components are proportional to the square of
wave amplitude and hence will be non-linear. In general, the drift force component has the
greatest magnitude, but this is dependent on hull shape and wave conditions (Arribas, 2007). In
Figure ‎1.4 the relative magnitude of the radiated and diffracted effects are shown conceptually.

4
FIGURE ‎1.4 - COMPONENTS OF ADDED RESISTANCE AND THEIR CONCEPTUAL RELATIVE SIGNIFICANCE AT
DIFFERENT WAVELENGTHS (HIROTA ET AL., 2005)

1.4 CALCULATION OF ADDED R ESISTANCE IN WAVES

Havelock (1942) was one of the first to estimate the added resistance in waves. By integrating
the longitudinal components of the pressure on the wetted surface of an oscillating ship hull, he
obtained a simple formula for the mean added resistance due to ship motions in heave and pitch.
He assumed that the incident wave system was undisturbed, hence no diffraction of waves, which
is called the Froude-Krylov approximation. The formulation requires no integration along the
ship length. Havelock’s expression for the added resistance, Raw, is,

Equation ‎1-1

k – wave number
Fa – amplitude of heave force
Za – heave amplitude
Ma – amplitude of pitch moment
- pitch amplitude
, – heave and pitch phase angle

The expression is not accurate due to exclusion of diffracted waves, coupled motions and viscous
damping. However, it shows that added resistance is partly due to the relative motion between
exciting waves and ship motions. It can be seen that the maximum of added resistance will be
obtained in resonance of vertical motions, thus poor sea keeping characteristics will induce
larger added resistance.

1.4.1 MOMENTUM AND ENERGY METHOD

Maruo (1957) also contributed to analytical calculation of the added resistance with a potential
flow solution. He used the equations for conservation of energy and momentum to derive his
method. The hull form was represented by a centre line distribution of singularities, and the
wave field potential consists of the potential associated with the original wave field and the
velocity potential of waves produced by the singularities.

This is a far field method, which is also called drift force method. Maruo (1960) developed the
basis of this method. He developed a general far-field theory to calculate the non-linear
hydrodynamic forces experience by a ship oscillating in incident waves without forward speed.

The method derives an energy and momentum balance on a control volume around the ship. The
velocity potential is divided into three parts, incident wave potential, diffracted wave potential

5
and radiated wave potential. The added resistance is then found from solving a boundary value
problem.

The incident wave potential is known, and solving the problem then becomes finding the
harmonic potential that satisfies a linearized free surface condition, a far-field radiation condition
and the ship hull boundary condition. (Arribas, 2007)

However, the diffraction effect is not included and it seems to over predict the added resistance
at low frequencies.

Joosen (1966) obtained a formula equal to Havelock (1942), except for an additional term for
coupled motions, by expanding Maruo’s results in an asymptotic series in terms of length to beam
ratio. He included forward speed by considering the encounter frequency, . The expression
also includes damping. The expression found can be written as,

Equation ‎1-2

, , – motion damping coefficients

– encounter frequency

He concluded that the added resistance due to waves is mainly caused by radiated waves. The
diffraction effects could be neglected, except for very small waves (Arribas, 2007). It also shows
best results for finer ship hulls with low Cb when compared to experimental tests (Arribas, 2007).

Kashiwagi (2009) calculated the added resistance of a ship with the modified version of Maruo’s
approach, using the Enhanced Unified Theory (EUT). In strip theory, 3D and forward speed
effects are ignored, but are incorporated in the EUT. In Kashiwagi et al. (2010) discrepancies
between experimental results and estimated values with the method are especially large for a
ship with forward speed in short waves.

1.4.2 INTEGRATED PRESSURE METHOD

This method is based on the classical hydrodynamic solution as is the one performed by Havelock
(1942). However, a more sophisticated method of calculating motions is applied with e.g. a strip
theory. This also enables one to calculate a small contribution of vertical motions due to the pitch
angles that produces a longitudinal force component.

When using strip theory, the longitudinal force must be estimated as a mean value for each
section. This is due to one of the approximations of strip theory dictating that no longitudinal
effects can be transferred between the strips. Boese (1970) obtained a mean value for a section
(strip), , at xb which is,

( ) Equation ‎1-3

– wave elevation
– vertical motion due to pitch
s – relative vertical motion amplitude
– phase angle of relative vertical motion

This is integrated over the waterline plane and results in

6
 Equation ‎1-4
∫( )

The contribution of the vertical motion is obtained by

Equation ‎1-5
Thus the total added resistance could be summed up

Equation ‎1-6
This method neglects the quadratic velocity term in Bernoulli’s equation as well as the term due
to the instantaneous wetted surface pressure. The method is also limited to head sea.

Faltinsen et al. (1980) derived a similar formula as Boese (1970). However, this derivation
included the quadratic velocity term as well as the pressure term arising from the instantaneous
wetted surface. The procedure of Faltinsen et al. (1980) is valid for any wave direction, in
contrast to Boese’s procedure. They also calculated the transverse drift force and yaw moment.
In the same article Faltinsen et al. (1980) derived an asymptotic formula for added resistance in
short waves. This will be commented upon in more detail later in this chapter.

1.4.3 RADIATED ENERGY METHOD

This method equates the work of added resistance to the energy contained in the damping waves
radiated away from the ship. This method was first applied by Gerritsma and Beukelman (1972).
The method is simple to calculate due to the fact that no boundary value problem needs to be
solved. It has proven to be accurate in predicting the added resistance in long waves (Wilson,
1985).

However, the effect of diffracted waves is only considered by correcting the ship motions with
one empirical equation. Thus this method cannot predict the added resistance in short waves
accurately.

1.4.4 ADDED RESISTANCE IN SHORT WAVES

The focus on the above mentioned methods have been on the energy dissipated through
interaction between the incident wave system and the radiated waves related to ship motions.
This effect is the dominating effect in the range of wavelengths around the ship length. However,
in the short wave length region the diffraction of waves is important. For short waves the ship
motions become small, and especially for blunt bows, the major part of the energy dissipation is
through reflection of incident waves.

The above-mentioned methods for calculating added resistance in waves have proven poor in
short waves (Wilson, 1985, Arribas, 2007). Special methods have been developed for added
resistance in short waves.

Fujii and Takahashi (1975) expanded Mauro’s method by considering added resistance in short
waves. The formula for drift force based on the wave reflection developed by Havelock (1942)
was used together with empirical corrections. Modifications and improvements on Fujii and
Takahashi’s formula have been done by Sakamoto and Baba (1986), Matsumoto et al. (1998) and
Ueno et al. (2001). However, these do not provide a significant difference from Faltinsen’s
asymptotic formula (Guo and Steen, 2010b).

Faltinsen et al. (1980) developed an asymptotic formula for the added resistance in short waves.
It uses the incident waves on an infinitely long plane wall to simulate the diffraction problem in

7
short waves. This method includes the wave reflection effect without consideration of ship
motions.

The asymptotic formula can be written as

∫̅ Equation ‎1-7

Where ̅ is the force per unit length normal to the hull given as

̅ ([ ] ) Equation ‎1-8
Equation ‎1-9

√ Equation ‎1-10

– wave amplitude
- angle between the tangent of the waterline and the centreline (x-axis)
- wave propagation direction with respect to the x-axis
L1 – non-shadow part of the water plane curve
- circular frequency of encounter
V – horizontal steady velocity parallel to the ship side
k – wave number

The assumptions that the vessel has vertical sides at the water-plane, the wave-induced ship
motion is negligible and the wave energy decays exponentially in depth enable the ship to be
replaced by a stationary vertical, infinitely long cylinder with the same cross section in the water-
plane as the ship.

Further, the application area of the method is limited to short waves, blunt ships and moderate
Froude numbers (FN <~0.2, (Faltinsen, 1990)). The method has been found to correctly predict
the added resistance when these formal conditions of the formula are satisfied. However, it is
very sensitive when the conditions are not met, and under-predicts the added resistance when
the ship sides are not vertical or the bow is not blunt (Steen and Faltinsen, 1998).

Guo and Steen (2010a) utilized a radiated energy method and the asymptotic formula of
Faltinsen et al. (1980) combined with an R-function to predict the added resistance on KVLCC2 in
a wider range of wavelengths. They compared the results to experiments conducted on a model
of KVLCC2. A new data processing approach in experiments was used due to unstable waves in
short waves causing difficulty in testing. They found that the radiated energy method seriously
under-predicts the added resistance in short waves, as mentioned earlier. The combined method
can predict the added resistance over a range of wavelengths. However, the method slightly
under-predicts the added resistance at lower Froude numbers, while it is predicting it well for
Froude numbers around 0.142-0.18.

1.4.5 CALCULATIONS OF ADDED RESISTANCE WITH COMPUTATIONAL FLUID DYNAMICS

(CFD)
With the increasing computational capacity, computational fluid dynamics is becoming more and
more available as a design tool.

Utilizing CFD approach relying on the resolution of the Reynolds Averaged Navier-Stokes (RANS)
equation can, as opposed to potential flow, handle effects of viscosity, wave dispersion,
nonlinearity and wave breaking (Choi and Yoon, 2009). Turbulence effects can be taken into
account through a turbulence model. A much-used approach to handle the free surface is Volume

8
Of Fluids (VOF). This has proved suitable for flow involving hull shape with section flare and
breaking waves (Azcueta, 2004).

Guo et al. (2011) did systematic validation and verification of numerical computation to
demonstrate that reliable numerical results can be obtained in calm water as well as head waves
on KVLCC2. A comparison between CFD results, the radiated energy method by Gerritsma and
Beukelman (1972), and experimental results show that the CFD results have the best agreement
with experimental results in most wavelengths. CFD is far better than the radiated energy
method in small wavelengths, while fairly better to predict the added resistance at the peak.

Guo et al. (2011) also studied the contribution of ship motions compared to diffracted waves on
the added resistance, a division proposed by Fujii and Takahashi (1975), by fixing the ship in CFD
calculations. It was found that for wavelengths shorter than 0.63Lpp, the influence of ship motions
on added resistance was negligible and the ship could be fixed. Increasing the wavelength
increases the difference in added resistance results between a fixed and free ship.

1.5 DISCUSSION OF METHODS

A calm water resistance method is used to identify changes in the calm water resistance due to
changes in the bow section.

For the calm water resistance case, several numerical tools based on potential theory are
available. Empirical methods only generates a statistical value based on a database of similar
ships and would not be able to capture small changes of the hull shape. To be able to capture the
changes of the bow shapes, one of the numerical solvers have to be used.

Due to the focus on large, slow ships in this thesis it is important to have a method that can
calculate the short wave added resistance as well as the long wave added resistance. The
radiation is expected to be less significant to these types of ships than for smaller ships. The
models used are, as an approximation, initially straight walled and blunt. The flare has been
removed. This enables the use of Faltinsen’s formula in short waves, which have shown
reasonably good results if the conditions are met. The formula is also easy to implement in a
computer code. In the radiation region several methods are applicable, an integrated pressure
method based on Faltinsen’s method (refer to chapter ‎1.4.2) or a radiated energy method based
on Gerritsma and Beukelman’s method (refer to chapter ‎1.4.3). A comparison and evaluation of
these methods will be done in chapter ‎5.1.3.

Another possibility is the use of CFD, more specifically RANS with VOF. This is precise for a wide
range of wavelengths and bows that are not straight. However, this requires extensive
computational time. Taking into account that due to short waves, the mesh will have to be very
fine, the computational effort required is significant. However, when investigating the efficiency
of unconventional bows CFD is needed to capture higher order effects that may occur when
having inclined surfaces above the waterline.

The next chapter will establish the chosen methods that are expected to be able to capture the
needs explained above.

9
2 METHODS
Methods to calculate the calm water and added resistance due to waves will be established in this
chapter. The methods will be verified in chapter ‎5.1 by comparing results with benchmark data
given in the literature. A method to implement wave statistics in evaluation of route performance
for the ships will also be established as well as a method to evaluate the efficiency of an
unconventional bow will also be established.

2.1 CALM WATER RESISTANCE

The calm water resistance needs to be established together with the added resistance for the
different bow shapes to be able to evaluate the total performance of the design on a typical
voyage.

The goal of the calm water resistance analysis is to establish a resistance curve as an input to the
speed-loss calculations done in ShipX.

Empirical methods have been avoided, as much as possible, when estimating the calm water
resistance in this thesis. The argumentation is that empirical methods are based on main
characteristics of the ship and will not be able to properly account for details in the design of the
bow. However, some methods have been used where it is assumed not to distort the results, and
when computational effort is either difficult or too time consuming.

The calm water resistance is decomposed into different contributions. Two major components
are the viscous resistance and the wave making resistance. The total resistance in calm water can
be decomposed as

CTs=CR+(1+k)(CFs+ CF) + CApps + CAAs+CDBs + CA Equation ‎2-1

CTs – Total resistance coefficient in ship scale

CR – Residual resistance coefficient
CFs - Skin friction coefficient in ship scale
CF – Hull roughness coefficient
CApps – Appendix resistance coefficient in ship scale
CAAs – Air resistance coefficient for ship
CDBs – Resistance coefficient of transom stern in ship scale
CA – Correlation allowance due to model test results

CA is the correlation allowance which is only added if CR is found from model tests. The residual
resistance is composed of several components. There is no standard decomposition of this
coefficient, which may contribute to difficulties in comparison of results. One major component
of the residual coefficient is the wave making resistance (CW), which again can be composed to
wave pattern resistance and wave breaking resistance (Larsson and Raven, 2010). The viscous
pressure resistance, form effect on pressure, is sometimes accounted for by the form factor, and
sometimes in the residual coefficient. This shows that the residual coefficient is not a well-
defined component, and one need to be aware of the definition in each case.

In this thesis Friendship-framework with Shipflow integrated has been used to evaluate the wave
resistance of the designs. Due to time constraints full CFD calculations of the viscous resistance
have not been done with Shipflow. Instead the ITTC ‘57 formula as well as other empirical
methods has been used to calculate the different contributions to viscous resistance from e.g.
skin friction coefficient, hull roughness and transom stern.

10
2.1.1 WAVE RESISTANCE
The Friendship-framework is a program that focuses on optimization of flow-related tasks. It is a
post-processing tool and provides methods for automatized optimization. It is set up to
collaborate with different CFD software, which provides the analyses of the case being worked on.
In this thesis the Friendship-framework is combined with Shipflow to calculate the wave
resistance of the designs which is a component of the residual resistance coefficient, C R.

XPAN is the solver used to calculate the wave resistance coefficient. It is a potential solver using a
Rankine source panel method. Shipflow calculates the wave resistance in two ways, by
transverse wave cuts and with pressure integration. With transverse wave cuts, Shipflow
calculates the energy from the wave pattern that radiates from the ship through a transverse
boundary behind the vessel. In the other method Shipflow calculates the pressure on the hull and
this pressure is integrated over the hull surface, giving the wave resistance.

There are advantages and disadvantages with both methods. The pressure integration is very
dependent on a good mesh, and in this case it seems like the low Froude number combined with
the standard mesh produces poor results. Therefore the wave resistance coefficient from the
wave cuts, which is less dependent on the mesh on the hull, is used to compare the different
designs. However, this is sensitive to the mesh size on the free surface, as waves are numerically
damped away from the vessel. Details of the results and verification can be seen in chapter ‎0.

The potential solver is not capable of capturing wave breaking, which can be a significant
component of the wave resistance on large blunt ships, and thus leads to an error in the
calculations.

2.1.2 VISCOUS RESISTANCE

The skin friction is directly proportional to the wetted surface of the ship, as this is defined as the
friction on a plate with the same wetted surface as the vessel. To account for increased velocity
over the hull surface due to the displacement of streamlines by the ship the form factor, , has
been applied. As mentioned in chapter ‎2.1 above, the form effect on pressure is sometimes
accounted for in the form factor. Thus giving a form factor that is .

2.1.2.1 P LATE FRICTION R ESISTANCE

The skin friction resistance coefficient is calculated according to the ITTC ’57 formula. It is
defined as

Equation ‎2-2

1.60E-03
1.55E-03
1.50E-03
1.45E-03
CF

ITTC '57 formula

1.40E-03
1.35E-03
1.30E-03
7.00E+08 1.70E+09 RN 2.70E+09

FIGURE ‎2.1 - ITTC-CURVE FOR FRICTION

11
2.1.2.2 F ORM F ACTOR
The form factor can be calculated by different empirical formulas. Such empirical formulas can
never represent anything else other than the average of ships used in the analyses. Thus using
such empirical methods could give a distorted result that does not represent the actual relative
resistance between different designs.

One empirical formula that takes into account both the form effect on pressure and friction is
Holtrop’s method for predicting the form factor. It is shown below, (Minsaas and Steen, 2008).

( )
Equation ‎2-3
( ) ( ) ( )

L=Lwl
CP – Prismatic coefficient
= Distance from the bow to the parallel mid ship.

CStern – Coefficient describing the shape of the stern. CStern=10 (U-shaped)

This formula uses the prismatic coefficient, CP, which describes the fullness of the ship towards
aft and bow. A large CP describes a ship with relatively large portion of the volume towards the
ends. However, it is not given that this formula will give a correct relation between the different
designs as it is very sensitive to main dimensions of the ship.

Prohaska’s method is a method used to establish the form factor from experiments, and
MARINTEK has developed one empirical formula as well, which does not include the form effect
on pressure.

2.1.2.3 H ULL R OUGHNESS

The hull roughness is also a significant contribution in resistance. It is calculated as an increase of
the skin friction. The hull roughness is dependent on the type of coating, amount of rust, fractures
in the coating, fouling etc. For ships in service the hull roughness resistance will increase during
the lifetime due to fouling of the hull. Thus the hull roughness allowance calculated here is a first
estimate. Results from measurements in wind tunnel, towing tank or cavitation tank have given
rise to an approximated formula shown below, (Minsaas and Steen, 2008).

Equation ‎2-4

H( ) – roughness
V – ship velocity (m/s)

When the roughness allowance according to Equation ‎2-4 is below zero, the roughness allowance
is taken as zero.

2.1.2.4 T RANSOM STERN

KVLCC2 has a wet transom stern where the flow is separated and creates a suction pressure. This
suction pressure arises as a result of the external flow, which carries away the dead water behind
the transom. On ships with a large transom and low Froude numbers this component may be
considerable. On KVLCC2 it is not very significant, but is still taken into account.

A formula based on experiments with projectiles in air has been made. When formulated using
the wetted surface area it becomes (Minsaas and Steen, 2008):

12
 √ Equation ‎2-5

This effect is not taken into account in the form factor due to the special dependence of CF.

2.2 ADDED R ESISTANCE IN WAVES

The added resistance is calculated using ShipX. ShipX is a tool developed at MARINTEK in
Trondheim. This tool has a plug-in (VERES) for calculating ship responses and motions based on
the low speed strip theory formulation by Tuck et al. (1970). The added resistance post
processor calculates the added resistance. The software gives a choice between strip theory and
pressure integration for calculating responses. Choosing to calculate added resistance based on
pressure integration gives a choice of methods in the long wave (radiation) region, either
Gerritsma and Beukelmann or a pressure integration approach based on the methods in
Faltinsen et al. (1980). Both approaches are combined with Faltinsen’s asymptotic formula for
added resistance in short wave (diffraction) region. Choosing ordinary strip theory leaves only
Gerritsma and Beukelmann’s method for calculation of added resistance in the radiation region.

To be able to combine the methods for added resistance due to ship motion induced wave
generation and the asymptotic formula for short waves the software use the method of Fujii and
Takahashi (1975), by multiplying the asymptotic formula with a correction factor that accounts
for finite draft and wavelength. A more detailed explanation can be found in the VERES Theory
Manual, (2010).

2.3 ADDED R ESISTANCE DUE TO WIND

Added resistance due to wind is not a focus in this thesis. The method used by Sunde (2011) is
also adapted in this thesis. In ShipX a standard wind coefficient for a tanker in loaded condition is
defined. It varies as a function of wind direction, which is set to the same as wave heading.

The wind speed is calculated by

√ Equation ‎2-6

HS – Significant wave height of the wave spectrum

The projected area is taken from calculations done by Sunde (2011) and is set to be Aproj = 920.8
m 2.

2.4 ROUTE SIMULATION

ShipX also calculates the involuntary speed-loss on a route, based on engine and propeller
characteristics, long term wave statistics, ship motions, calm water and added resistance from
VERES.

It takes into account possible reduction in propeller efficiency as the vessel encounter waves due
e.g. change in relative submergence of the propeller and corrects the open water diagram. It can
also take into account change in engine efficiency due to change in loading.

The calculation of ship motions and added resistance is done in VERES and the results are used
as input in the speed-loss calculations.

13
To calculate the speed-loss in irregular waves the open water diagram is corrected over regular
waves. The wave-spectrum realization is cut into successive regular wave parts. Each regular
wave has a given probability of occurrence based on Lounget-Higgins joint probability density
function applicable to wave amplitude and period.

The speed-loss is calculated for each wave component. The speed-loss in the given sea state is
calculated as a weighted average of the speed-losses for each component where the joint
probability density is the weighting factor for the wave component.

To find the speed-loss on a given route the weighted average of the speed-loss for each sea-state
is calculated. The weighting factor is based on the weighted scatter diagram described in
chapter ‎4.2.1, containing information on the probability of occurrence for each sea state on the
route.

2.5 DESIGNING THE MODELS

Variations of the bow have to be designed. The tool used for this is the Maxsurf suite by Formsys.
The suite contains tools to do hull design with Maxsurf Pro, and other tools can be used for
stability calculations, sea keeping calculations and resistance calculations to mention some.

In this thesis only the hull design features is used. The models are changed according to the
methods described in chapter ‎3. More information on the software can be found here:
https://www.formsys.com/maxsurf.

14
3 CHANGING BOW DESIGN
Evaluating the trade-off between calm water resistance and added resistance due to waves when
changing the sharpness of the bow is not as straight forward as it might seem. The design is
based on a finished design of a VLCC. Thus the base design is already designed for certain
requirements and speed. Such requirements could be a ship owner’s demand to load capacity and
length due to size restrictions in ports or on special legs of voyages in canals etc.

3.1 RESISTANCE CONSIDERATIONS

When changing the bow design, it is important to evaluate all the aspects of the resistance. One
can risk decreasing the efficiency of the design due to focusing on only one parameter. This
section will look into the different resistance components and how they presumably will change
with entrance angle. The resistance components that are assumed to not or to a small degree be
influenced by the changes in the bow are not discussed, e.g. the wet transom stern resistance.

The friction component is ruled by the wetted surface of the underwater hull as well as the hull
surface roughness. Thus a design that increases the wetted surface will increase the skin friction.
It is not dependent on the shape of the hull, and thus it is difficult to avoid, other than minimizing
the wetted surface. A sharp bow will by mathematical considerations have a larger wetted
surface compared to a blunt bow, of the same displacement. The friction resistance is one of the
major resistance components and a percentage change thus contributes more to an increase or
decrease of the total resistance than other components. However, considering the large wetted
surface area on KVLCC2, the change in wetted surface due to changes in the bow may not be
significant.

The form factor is dependent on the shape of underwater hull. As explained in ‎2.1 it incorporates
the increase of velocity over the hull due to the displacement of the streamlines as well as the
form effect on the pressure. A longer, more slender bow increases the overall slenderness of the
vessel, and thus the form factor should be decreased for a longer bow.

The wave making resistance is small for the large, slow ships compared to viscous resistance due
to the low Froude number. It is very dependent on the shape of the hull. In theory the wave
resistance could potentially increase drastically with changes in the bow shape due to e.g.
interaction effects between the bow wave system and the fore shoulder wave system. Due to the
low significance of this component, it is not a major concern. However, it should be monitored
because it may be relatively simple to avoid such effects, thus avoiding unnecessary increase in
the resistance.

The added resistance due to reflection of waves in head seas will decrease when the bow is
sharpened, as found in Guo and Steen (2010b). The degree of change in the bow is a trade-off
between the ratio of time spent in calm water and wave conditions and the severity of the wave
conditions. If the vessels encounters short waves most of the time, a sharper bow may be optimal.
However, if the encountered waves are in the radiation regime the majority of the operating time
a sharper bow is expected to be less, as the motion characteristics are most important in this
range. The sharpness also rules the amount of reduction of added resistance in waves. Thus, a
sharper bow will in theory have less added resistance, but the trade-off with calm water
resistance will restrict an extremely long bow.

3.2 LENGTH AND DISPLACEMENT CONSIDERATIONS

Sharpening of the bow shape can be done in basically two ways. This is schematically shown in
Figure ‎3.1 below. The black line represents a half-circle shaped bow. The blue line is a sharpening
of the bow when keeping the length fixed. The volume is removed from the shoulders. This will

15
require the displacement to decrease, which in turn will require the available cargo space to
decrease. This might be a problem for the ship owners wanting to carry as much cargo as
possible for a given length. This way of sharpening the bow will possibly decrease the wetted
surface, considering skin friction resistance, though sacrificing volume.

FIGURE ‎3.1 - SCHEMTAICALLY OPTIONS OF CHANGING THE BOW SHAPE. BLACK LINE IS ORIGINAL BOW SHAPE,
BLUE LINE IS DECREASED DISPLACEMENT AND RED LINE IS LENGTHENED SHIP.

Increasing the length as shown in the red line will increase the displacement, of the ship.
However, this may cause issues regarding length requirements in ports or on certain parts of a
route such as canals.

Keeping the displacement fixed can be done by shifting volume from the shoulders to the front of
the bow as shown in Figure ‎3.2. Here the volume in 1 and 2 are shifted to 3. This will maintain the
displacement. It is basically a combination of the extremes shown in Figure ‎3.1. However, the
length will in this case also necessarily be increased. Regarding resistance, the wetted surface to
volume ratio of this transformation will increase.

When changing the bow shape from the base-case, hydrostatic values may also change. One
important hydrostatic value is the Longitudinal Centre of Buoyancy (LCB). The original design
has an LCB that is coordinated with the Longitudinal Centre of Gravity (LCG). Shifting the LCB
will induce a change in trim on the ship.

When designing alternatives the LCB is monitored closely and an effort is put into keeping it as
constant as possible. However, shifting volume forward will induce some change in the LCB and
an assumption has to be made that the trim can be compensated by shifting weights. Due to the
sheer size of KVLCC2 the change in percentages of LPP is small as can be seen if one compares
KVLCC2A and KVLCC2C in Table ‎3.1 and Table ‎3.3.

16
FIGURE ‎3.2 - COMBINATION OF THE TWO DESIGNS. THE VOLUME MOVED FROM 1 AND 2 IS EQUAL TO VOLUME 3

The original KVLCC2 is designed with a bulb extending 8 meters in front of the fore
perpendicular. Regarding the length of the ship as L=L PP+Lbulb it is possible to move volume from
the shoulders and forward over the bulb as seen in Figure ‎3.3.

FIGURE ‎3.3 - THE HATCHED AREA CAN BE USED TO MOVE VOLUME FORWARD AND MAKING THE BOW SHARPER

17
3.3 THE DESIGNS
Several designs have been made for this thesis. They represent different waterline curves and
approaches to changing the bow. A short explanation of the design and characteristics are
presented below.

3.3.1 KVLCC2A
KVLCC2 is the second variant of the MOERI tanker that has been used as an academic standard
within CFD validation. The additional notation A means, in this thesis, that it is the original design
with bulb. This has been used for verification of the methods used to analyse the calm water and
added resistance. To fulfil the requirement in Faltinsen’s method for added resistance in short
waves, the flare in the bow has been removed such that the sides are more or less straight in the
waterline. The body plan of the ship can be seen in Figure ‎3.4.

The iges-file provided on Simman2008 (2008) had many surfaces that produced poor offsets in
Shipflow and thus it had to be redesigned in Maxsurf to get at smooth surface. Due to this, the
main particulars of this ship, shown in Table ‎3.1 are slightly different from the given in Table ‎1.1.

FIGURE ‎3.4 - KVLCC2A BODY BLAN

TABLE ‎3.1 - MAIN DIMENSIONS KVLCC2A

Main Particulars KVLCC2A

Lpp (m) 320.0
Lwl (m) 325.5
Bwl (m) 58.0
D (m) 28.0
T (m) 20.8
Displacement ∇ (m3) 312948.6
2
S excl. rudder (m ) 27663.4
CB 0.797
CM 0.999
LCB (%). fwd+ from amidship 3.46
LCB (m) from AP 171.3

18
3.3.2 KVLCC2B
KVLCC2B in this thesis is similar to KVLCC2B used in Guo (2011). However, due to lack of a 3D
model the version in this thesis was designed in Maxsurf based on the description and main
dimensions given in Guo (2011).

The main concept of this design is the straight sides all the way down with a small bilge radius.
The bulb is removed and it has the same curvature in the waterline as KVLCC2A. The wetted
surface area of this is considerably larger and thus is expected to produce poor overall results. It
is used in the analysis to study the effect of wetted surface area on the overall performance.

FIGURE ‎3.5 - KVLCC2B BODY PLAN

TABLE ‎3.2 - MAIN DIMENSIONS KVLCC2B

Main Particulars KVLCC2B

Lpp (m) 320
Lwl (m) 325.5
Bwl (m) 58
D (m) 28
T (m) 20.8
3
Displacement ∇ (m ) 314373.1
2
S excl. rudder (m ) 28146.4
CB 0.8
CM 0.999
LCB (%). fwd+ from amidship 3.559
LCB (m) from AP 171.6

19
3.3.3 KVLCC2C
KVLCC2C was also produced by Guo (2011). It is based on line drawings provided by Professor
Sverre Steen. It has a sharper entrance angle than KVLCC2A and KVLCC2B. It is designed by the
principle of moving volume from the shoulders to fill in the volume above the bulb, thus the LPP is
lengthened by 8 m and the CB is slightly reduced. However, the overall length is maintained. The
bow is slightly trimmed down below the waterline to keep the displacement and LCB as constant
as possible.

FIGURE ‎3.6 - KVLCC2C BODY PLAN

TABLE ‎3.3 - MAIN DIMENSIONS KVLCC2C

Main Particulars KVLCC2C

Lpp (m) 328.0
Lwl (m) 333.5
Bwl (m) 58.0
D (m) 30.0
T (m) 20.8
3
Displacement ∇ (m ) 313049.4
S excl. rudder (m2) 27787.7
CB 0.776
CM 0.998
LCB (%). fwd+ from
2.169
amidship
LCB (m) from AP 171.2

20
3.3.4 KVLCC2D
KVLCC2D is a blunter version of KVLCC2A, however, without bulb. It is made to get even more
extreme results from the analyses. KVLCC2A is blunt, but has a sharp angle in the bow, thus this
is made to get a more continuous curve at the bow. The length of 320 m is kept constant and the
displacement is increased and thus the CB is also larger. The lines under the waterline are
somewhat slimmer than KVLCC2A to reduce the displacement and wetted surface that increases
due to the fattening of the bow. The LCB is further forward from the Aft Perpendicular (AP) than
for KVLCC2A and an assumption is made that it is possible to move weights forward due to the
increased volume in the bow.

FIGURE ‎3.7 - KVLCC2D BODY PLAN

TABLE ‎3.4 - MAIN DIMENSIONS KVLCC2D

Main Particulars KVLCC2D

Lpp (m) 320
Lwl (m) 325.5
Bwl (m) 58
D (m) 28
T (m) 20.8
Displacement ∇ (m3) 314992.6
S excl. rudder (m2) 27691.8
CB 0.8
CM 0.999
LCB (%). fwd+ from
3.647
amidship
LCB (m) from AP 171.9

21
3.3.5 KVLCC2E
KVLCC2E is an elongated version of KVLCC2C, without any bulbous curvature in the bow. It is on
the other side of the extremes compared to KVLCC2D. The bow is elongated by 8 m to 336 m
compared to KVLCC2C. Due to the elongation of the bow, the C B is reduced and the LCB is moved
forward.

FIGURE ‎3.8 - KVLCC2E BODY PLAN

TABLE ‎3.5 - MAIN DIMENSIONS KVLCC2E

Main Particulars KVLCC2E

Lpp (m) 336
Lwl (m) 341.5
Bwl (m) 58
D (m) 28
T (m) 20.8
3
Displacement ∇ (m ) 314026.6
S excl. rudder (m2) 27982.8
CB 0.774
CM 0.998
LCB (%). fwd+ from
1.212
amidship
LCB (m) from AP 172.1

3.3.6 WATERLINES
In Figure ‎3.9 below the waterlines of the different designs are shown as well as the half angle of
entrance in the bow. As seen in the figure, KVLCC2E is longer and sharper, and should thus
reduce the added resistance, at least the diffraction component in head seas. However, due to the
increased wetted surface and frictional resistance one can expect a decreased benefit.

It can be seen by the tables above that KVLCC2D has more volume than KVLCC2A, but the wetted
surface is more or less the same. As mentioned this was achieved by slightly slimming down the
underwater hull at the bow compared to KVLCC2A. The removal of the bulb also influences to
reduce the wetted surface, as the bulb has a high surface area to volume ratio.

In the figure the lines of KVLCC2A and KVLCC2B coincide.

22
FIGURE ‎3.9 - WATERLINES OF THE DESIGNS

3.3.7 SECTIONAL AREA CURVE

The sectional area curve for KVLCC2A and a detail of the bow area of all the designs are shown in
Figure ‎3.10 and Figure ‎3.11.

1400

1200

1000
Area (m2)

800

600 KVLCC2A
400

200

0
-60 40 140 240 340
Position (m)

FIGURE ‎3.10 - SECTIONAL AREA CURVE OF KVLCC2A

KVLCC2B has more volume between 260 m and ~300 m due to the straight sides all the way
down. It is also without bulb, as all the designs except KVLCC2A which makes it shorter than
KVLCC2A. Both KVLCC2C and KVLCC2E have some volume cut from the shoulders between 270
m and 310 m. However, these are longer in the waterline and thus this volume is moved forward.
KVLCC2D is the bluntest of the designs and this is seen by the steeper ending of the sectional area
curve. It has more area than KVLCC2A between 250 m and 315 m because the shoulders have
been moved forward and thus has a higher displacement.

23
 1200
1100
1000
900
800
KVLCC2A
700
Area (m2)

KVLCC2B
600
KVLCC2C
500
KVLCC2D
400
KVLCC2E
300
200
100
0
230 240 250 260 270 280 290 300 310 320 330 340
Position (m)

FIGURE ‎3.11 - SECTIONAL AREA CURVE OF THE BOW AREA OF ALL DESIGNS

3.3.8 EXPECTED PERFORMANCE

The relative expected performance on a route between the different designs regarding key
parameters frictional resistance and added resistance can be seen in Figure ‎3.12. The frictional
resistance is the major component of the calm water resistance and is thus used as a parameter
here. This chart shows the relative performance in added resistance based on bow shape, and the
frictional resistance based on wetted surface. It is only a graphical presentation of the personal
perception of level of performance and not weighted by the significance of added resistance and
frictional resistance on the total performance. The difference in added resistance is based mostly
on the diffraction effect as the motions and radiation is more difficult to intuitively range.
More frictional resistance -->
(Larger wetted surface)

KVLCC2A
KVLCC2B
KVLCC2C
KVLCC2D
KVLCC2E

More added resistance -->

(Blunter bows)

FIGURE ‎3.12 - RELATIVE PERFORMANCE BETWEEN DESIGNS REGARDING FRICTIONAL RESISTANCE AND ADDED
RESISTANCE

Figure ‎3.12 shows that the expected performance of KVLCC2B is poor both with regards to added
resistance and frictional resistance, due to the blunt bow and straight sides.

24
KVLCC2D is expected to perform well with regard to frictional resistance, however, the worst to
added resistance.

KVLCC2C has bigger a wetted surface than KVLCC2A, but not nearly as much as KVLCC2E or
KVLCC2B. The increase in wetted surface due to the longer bow, is not as big as one might expect
compared to KVLCC2A and KVLCC2E. This may be ascribed to the fact that KVLCC2A has a bulb,
and thus some area in front of the fore perpendicular that does not represent an increase in the
wetted surface area when designing KVLCC2C. The added resistance of KVLCC2C is expected to
be good. Overall the vessel is a compromise between KVLCC2A and KVLCC2E.

KVLCC2E is the extreme design with regard to added resistance. It has the worst performance in
frictional resistance of the naturally shaped bows, i.e. excluding KVLCC2B which has a large
wetted surface due to the inconvenient design.

It is difficult to give an expected best performer as it depends on the relative significance

between the different resistance components, and for added resistance the severity of the sea
states the vessel will encounter.

25
4 ROUTES AND WAVE STATISTICS
The calm water conditions most ship designs are optimized for is a rare event for the large ocean
going vessels. To be able to get a realistic evaluation of the performance of ships, the operating
profile should be taken into account. Due to time restriction, a full operational profile will not be
evaluated in this thesis, which is recommended if a design study is carried out. In this thesis only
the fully laden case with 100% DWT utilization will be investigated. This chapter will establish
typical routes and wave statistics for these.

4.1 ROUTES
The report on Triality (2010), by DNV, considers the main trade routes for VLCC’s world-wide.
They point out that most VLCCs originate in the Arabian Gulf and have three main markets; East
Asia, US and Europe. Based on this, two routes have been chosen as a typical voyage for a VLCC in
this thesis.

In addition a route across the North Atlantic, which hosts some of the most severe sea conditions
in the world, is chosen to get a variation of characteristic sea states.

Even though KVLCC2 is designed as a VLCC, the size and shape of the hull is close to a typical
large bulk carrier. Therefore one bulk route has been chosen in this thesis.

Typical bulk routes are dependent on the type of trade that is being considered. Cape size vessels
are focused on long haul iron ore and coal trade routes (Genco, 2012). Typical iron ore trade
routes are from Port Hedland in Australia or Brazil to China with e.g. Qingdao as discharge port.
Iron ore supply from Australia and Brazil accounted for 64.4% of the imported iron ore to China
in 2009 (China-Daily, 2010). Due to lack of wave statistics in the Java Sea, Banda Sea and Arafura
Sea only the route from Brazil to China has been chosen as a bulk route. The routes can be seen in
Figure ‎4.1.

4.1.1 ARABIAN GULF – GULF OF MEXICO

The first route originates in the Arabian Gulf. The origin port is typically Ras Tanura, one of the
largest ports in the world for crude oil (Saudi-Aramco, 2012). The voyage goes south down and
around the Cape of Good Hope, across the South Atlantic Ocean and ends in the Gulf of Mexico.
Louisiana Offshore Oil Port (LOOP) is the largest point of entry for crude oil coming in to the U.S.

4.1.2 ARABIAN GULF - JAPAN

The other route and a route that is expected to grow in the future is the route from the Arabian
Gulf to the Far East. Ras Tanura is again the origin port. The destination port is Chiba, Japan. This
is the largest port in Japan and handles large crude oil tankers.

4.1.3 WESTERN EUROPE – EAST COAST, NORTH AMERICA

North America, including west and east coast, imported about 682 million tonnes of oil in 2005.
In 2004 the seaborne trade from the North Sea to North America was 46 million tonnes of a total
seaborne trade from the North Sea of 62 million tonnes (Stopford, 2009). Mongstad in Norway is
one of the largest export ports of crude oil in the North Sea and is used as loading port, while
New York is used as a discharge port on the East Coast of North America.

4.1.4 BRAZIL – CHINA

Vale S.A. is a Brazilian mining company, second largest in the world, and the largest producer of
iron ore. The company is one of the largest logistics operators in Brazil and owns the port of
Tubarão, which is the largest iron ore export port in the world (Løvstad, 2008). The port of
Qingdao in China has a terminal for handling large iron ore bulk carriers.

26
4.2 WAVE STATISTICS
This chapter will briefly describe the basics of wave statistics and establish the wave statistics for
the typical routes chosen in the last section.

The wave conditions on a location will vary significantly over time. Even during hours the wave
conditions can change. Thus to evaluate the vessels for regular waves will have no value when
trying to estimate the best performance in a realistic scenario other than establishing the
characteristics of the vessel responses in waves over a range of wave periods. Irregular waves,
waves of different periods and heights, are the case in the real world and often the waves are also
spread out in direction, making the surface chaotic.

Typically, wave characteristics are measured by buoys or reported by ships. The measurements
of buoys are used to establish wave spectra. These spectra contain information about the energy
distribution over wave frequency and are more or less specific for a certain ocean area and
several spectra have been defined throughout the years. They are defined as an energy
distribution over period or frequency given by certain spectral parameters like the significant
wave height and a characteristic period (HS and TZ for Pierson-Moskowitz spectrum). Since the
waves in real life are spread out over the direction, the energy is too. A directional wave
spectrum describes not only how the energy varies with wave period, but also how it varies with
direction.

Short term statistics utilizes these spectra to describe the sea surface within a certain period of
time, where the parameters can be assumed constant, called a sea state. A widely used time
period is 3 hours. Thus one assumes that the sea state is constant for 3 hours. However, during a
year, or the lifetime of the vessel, the vessel will encounter a wide range of sea states that can be
described by changing the characteristic spectral parameters. However, some sea states appear
more often than others in an area. Thus some sea states have a higher probability of occurrence
when thinking statistics. To utilize long term statistics a probability distribution of the sea states
is thus needed. This will be described in the next chapter.

4.2.1 LONG TERM STATISTICS

The long term statistics are defined by a scatter diagram. This diagram shows the probability of
occurrence of a combination of significant wave height and a characteristic period. In this thesis
it is the zero crossing period, TZ. An example of the scatter diagrams used in this report can be
seen in ‎Appendix A.

To generate the needed scatter diagrams for the routes RouteSim by DNV is used. It contains
wave data for specified zones around the world.

The routes have to be defined with area number and the length ratio of the route segment in the
specified area. In Figure ‎4.1, the map of the predefined areas in RouteSim is shown. The red lines
in the map are the VLCC routes and the yellow are the iron ore bulk route described above. The
green is the North Atlantic route.

RouteSim contains information about scatter diagrams in each zone. These will be extracted and
weighted with respect to the length ratio of each area and a weighted scatter diagram for the
total route will be used as input to ShipX. The input to RouteSim is shown in Table ‎4.1. The area
number in Table ‎4.1 corresponds to an area on the map in Figure ‎4.1. The ratio is the length in
each area divided by the total length of the route in percentage. Table ‎4.2 shows the
corresponding length in nautical miles.

27
FIGURE ‎4.1 - MAP OF PREDEFINED AREAS AND ROUTES. RED LINES IS VLCC ROUTES AND YELLOW IS THE IRON
ORE BULK ROUTE. THE GREEN IS FROM MONGSTAD TO NEW YORK

TABLE ‎4.1 - AREA AND RATIOS USED AS INPUT TO ROUTESIM

Ras Tanura Ras Tanura Tubarão - Mongstad -

- LOOP - Chiba Qingdao New York
Area Ratio Area Ratio Area Ratio Area Ratio
Number (%) Number (%) Number (%) Number (%)
32 4.09 29 9.86 28 2.39 9 29.63
38 3.43 38 6.42 40 7.26 11 9.09
39 3.58 39 6.70 41 5.34 15 35.02
47 12.46 40 11.73 61 11.21 16 11.03
48 1.99 41 9.00 62 8.79 23 7.62
50 5.28 50 12.08 69 8.21 24 7.62
56 8.13 60 10.29 74 5.88
59 6.04 61 17.81 75 6.62
60 6.04 62 16.11 76 11.67
66 4.09 84 10.14
67 5.14 85 4.99
68 7.13 89 6.37
75 9.62 90 11.12
84 8.18
85 4.89
90 9.91

TABLE ‎4.2 - TOTAL DISTANCES (SOURCE: HTTP://WWW.SEA-DISTANCES.COM/)

Route Total distance (nm)

Ras Tanura - LOOP 12299
Ras Tanura - Chiba 6608
Mongstad - New York 3365
Tubarão - Qingdao 11086

28
In order to do long term statistics calculations in ShipX assumptions on the wave spectrum for
each route have to be made. ShipX gives a choice between Pierson-Moskowitz spectrum,
JONSWAP spectrum and Torsethaugen spectrum.

The JOint North Sea WAve Project (JONSWAP) spectrum is based on measurements in the North
Sea. The peak in the spectrum is more pronounced than in the Pierson-Moskowitz spectrum. This
is a characteristic of the waves in the North Sea.

The Pierson-Moskowitz spectrum describes a fully developed sea. This means that the wind has
been steady long enough for the waves to come to a sort of equilibrium with the wind. The
spectrum was developed from measurements in the North Atlantic during 1964.

The Torsethaugen spectrum is a double peak spectrum that has frequently been used for design
purposes at the Norwegian Continental Shelf, both in connection with numerical analyses and
model tests. The spectrum was established by fitting two JONSWAP shaped models to average
measured spectra from the Norwegian Continental Shelf (Torsethaugen and Haver, 2004).

Due to the limited availability of area specific spectra in ShipX, a simplification is made, and only
one type is assumed for the whole route. The most general for open ocean areas is the Pierson-
Moskowitz spectrum and this one is thus applied on all routes.

4.2.2 VALIDITY OF FALTINSEN’S FORMULA

In Faltinsen (1990) it is stated that there for a two parameter Pierson-Moskowitz spectrum is no
significant energy for

Equation ‎4-1
And by using the dispersion relation and requirements for short waves it is found that long term
statistics in combination with Faltinsen’s formula for added resistance in short waves can only be
used when

( ) Equation ‎4-2

This requirement gives that for a vessel of 320 m TZ<5.14 s.

4.2.3 SCATTER DIAGRAMS

The weighted scatter diagrams are presented below. There are no specified directions in these
diagrams as this is not provided by RouteSim, they are so-called omnidirectional. Hence, a
simplification has been made to be able to utilize the information in the wave statistics.

In the input file to ShipX the period utilized has to be specified, and ShipX then computes other
spectrum periods from this if needed. The scatter diagrams are given by HS and TZ, and thus TZ
will be specified in the input file of ShipX.

In Figure ‎4.2 the weighted scatter diagram for the route from Ras Tanura to Chiba is shown. The
floor axes show HS and TZ and the vertical axis shows the probability of the occurrence for a
combination of HS and TZ. As seen the probability of a sea state with a HS>6.5 m or TZ>10 s is very
low.

In Figure ‎4.3 the Cumulative Distribution Function (CDF) is shown. The cumulative distribution
function used is defined as

29
 ̂ Equation ‎4-3

n – total number of observations

– number of observations lower or equal to (TZ)

6-7
5-6
4-5
3-4
7 2-3
6 1-2
0-1
5
Probability (%)

4 0.5
2
3 3.5
5
2 6.5
8
1 9.5
11 Hs (m)
0 12.5
2 4 6 14
8 10 12
14 16
Tz (s)

FIGURE ‎4.2 - WEIGHTED SCATTER DIAGRAM FROM RAS TANURA TO CHIBA

By using the requirement given by Equation ‎4-2 we can see from Figure ‎4.3 that on the route
from Ras Tanura to Chiba about 50% of the sea states are below TZ=5.14 s. This shows that a
large portion of the encountered waves are within the short wave assumption of Faltinsen’s
formula and thus the diffraction effect cannot be neglected, and will be a significant contribution
to the speed-loss. Corresponding values for the other routes are shown in Table ‎4.3.

TABLE ‎4.3 - PERCENTAGE OF SEA STATES WITHIN THE LIMITATIONS OF FALTINSEN'S FORMULA FOR ADDED
RESISTANCE IN SHORT WAVES

Route ̂
Ras Tanura to LOOP ~20%
Ras Tanura to Chiba ~50%
Mongstad to New York ~11%
Tubarão to Qingdao ~20%

30
 0-10 10-20 20-30 30-40 40-50 50-60 60-70 70-80 80-90 90-100

16

14

12

10

Tz (s)
8

2
0.5 1.5 2.5 3.5 4.5 5.5 6.5 7.5 8.5 9.5 10.5 11.5 12.5 13.5 14.5

Hs (m)

FIGURE ‎4.3 - THE CUMULATIVE DISTRIBUTION FUNCTION IN PERCENT – RAS TANURA – CHIBA

From Table ‎4.3 it can be noted that the route across the North Atlantic is the one with least sea
states below the criteria for Faltinsen’s formula. This is as expected, as the North Atlantic is
known to be one of the harsher oceans in the world. This means also that added resistance due to
diffraction is of less significance in this area compared to the other routes and that the sharper
bows may be less effective on this route.

The scatter diagrams for the rest of the routes can be seen below. It can be noted that the two
routes departing from Ras Tanura have about 90% of the sea states below HS m, whilst
Mongstad – New York being the extreme case with 90% of the sea states below H S , with
just ~70% under HS m (Figure ‎4.7).

31
 4-4.5
4.5 3.5-4
4 3-3.5
3.5 2.5-3
Probability (%)

3 2-2.5
1.5-2
2.5 1-1.5
2 0.5-1
1.5 0.5
1 3
0.5 5.5
0 8
2 4 Hs (m)
6 8 10 10.5
12 14 13
Tz (s) 16

FIGURE ‎4.4 – WEIGHTED SCATTER DIAGRAM FROM RAS TANURA TO LOOP

0-10 10-20 20-30 30-40 40-50 50-60 60-70 70-80 80-90 90-100

16

14

12

10
Tz (s)

2
0.5 1.5 2.5 3.5 4.5 5.5 6.5 7.5 8.5 9.5 10.5 11.5 12.5 13.5 14.5
Hs (m)

FIGURE ‎4.5 – THE CUMULATIVE DISTRIBUTION FUNCTION IN PERCENT – RAS TANURA TO LOOP

32
 4-5
3-4
2-3
5
1-2
4 0-1
Probability (%)

2
0.5
1 3
5.5
0 8
2 Hs (m)
4 6 10.5
8 10
Tz (s) 12 14 13
16

FIGURE ‎4.6 – WEIGHTED SCATTER DIAGRAM FROM MONGSTAD TO NEW YORK

0-10 10-20 20-30 30-40 40-50 50-60 60-70 70-80 80-90 90-100

16

14

12

10
Tz (s)

2
0.5 1.5 2.5 3.5 4.5 5.5 6.5 7.5 8.5 9.5 10.5 11.5 12.5 13.5 14.5

Hs (m)

FIGURE ‎4.7 - THE CUMULATIVE DISTRIBUTION FUNCTION IN PERCENT – MONGSTAD TO NEW YORK

33
 4-5
3-4
5
2-3
Probability (%)

4 1-2
3 0-1

2
0.5
1 3
5.5
0 8 Hs (m)
2 4 6 8 10 10.5
12 14 13
Tz (s) 16

FIGURE ‎4.8 - WEIGHTED SCATTER DIAGRAM FROM TUBARÃO TO QINGDAO

0-10 10-20 20-30 30-40 40-50 50-60 60-70 70-80 80-90 90-100

16

14

12

10
Tz (s)

2
0.5 1.5 2.5 3.5 4.5 5.5 6.5 7.5 8.5 9.5 10.5 11.5 12.5 13.5 14.5
Hs (m)

FIGURE ‎4.9 - THE CUMULATIVE DISTRIBUTION FUNCTION IN PERCENT –TUBARÃO TO QINGDAO

34
5 ANALYSIS AND RESULTS
This chapter describes the set up and program of the different analyses and present the results.
Verification of the methods used for Shipflow is performed and presented. Three parts have been
evaluated, the calm water resistance and the added resistance in waves, which are the input to
the evaluation of designs on typical routes.

5.1 VERIFICATION
MARINTEK have done experiments on KVLCC2 in relation with the work of Guo (2011).
Resistance curves have been developed and these results will be used to verify the calculations
done with Shipflow.

The original model test results can be seen in Table ‎5.2. Characteristics of the model are given in
the table below.

TABLE ‎5.1 - MAIN DIMENSIONS OF SHIP AND MODEL USED IN MARINTEK EXPERIMENTS

Model scale 58
Ship Model
Lpp (m) 320 5.517
Lwl (m) 325.503 5.612
Bwl (m) 58.021 1
T (m) 20.8 0.359
2
S excl. rudder (m ) 27800.68 8.264
S transom stern 13.48 0.004
3
Displacement ∇ (m ) 312677.38 1.603
CB 0.796 0.796

Water density 1025 998.82

Kinematic viscosity 1.19E-06 1.10E-06

TABLE ‎5.2 - RESULTS FROM MARINTEK EXPERIMENTS, FORM FACTOR K = 0.2021

Model
FN RNm CTm CFm CAppm CBDm CR
0.101 3.80E+06 5.81E-03 3.58E-03 0.00E+00 5.00E-06 1.50E-03
0.121 4.56E+06 5.23E-03 3.46E-03 0.00E+00 5.00E-06 1.07E-03
0.143 5.39E+06 5.34E-03 3.36E-03 0.00E+00 5.00E-06 1.31E-03
0.157 5.91E+06 5.29E-03 3.29E-03 0.00E+00 5.00E-06 1.33E-03
0.161 6.07E+06 5.14E-03 3.28E-03 0.00E+00 5.00E-06 1.19E-03
0.181 6.83E+06 5.54E-03 3.21E-03 0.00E+00 5.00E-06 1.67E-03

Ship
FN RNs CR CFs CVs CBDs CApps CTs
0.101 1.56E+09 1.50E-03 1.45E-03 1.87E-03 8.00E-06 0.00E+00 3.25E-03
0.121 1.87E+09 1.07E-03 1.42E-03 1.87E-03 8.00E-06 0.00E+00 2.83E-03
0.143 2.21E+09 1.31E-03 1.39E-03 1.87E-03 8.00E-06 0.00E+00 3.06E-03
0.157 2.43E+09 1.33E-03 1.38E-03 1.87E-03 8.00E-06 0.00E+00 3.08E-03
0.161 2.49E+09 1.19E-03 1.37E-03 1.87E-03 8.00E-06 0.00E+00 2.94E-03
0.181 2.80E+09 1.67E-03 1.35E-03 1.87E-03 8.00E-06 0.00E+00 3.42E-03

35
The division of the resistance components from MARINTEK is that the viscous pressure
resistance is taken into CR. Thus the form factor used by MARINTEK is k=kf (refer to
chapter ‎2.1.2). In this thesis, the viscous pressure resistance is taken into account by the form
factor, thus the MARINTEK form factor is too low.

To recalculate the resistance coefficients with a new form factor the following procedure is used.

The new residual resistance coefficient is calculated from the model scale results by

Equation ‎5-1

k – the new form factor

In full scale the skin friction coefficient is not altered from the original results, but the viscous
resistance has to be recalculated with the new form factor. Below,
calculations with two different form factors have been performed. In Table ‎5.3 a form factor of
k=0.391 is applied, taken from numerical calculations in Steen et al. (2010) and a form factor
calculated by Holtrop’s method of k=0.3435. The residual resistance is now decreased and the
viscous resistance increased. The form factors used can be seen in the table. The hull roughness
is calculated by Equation ‎2-4.

TABLE ‎5.3 - RESULTS FROM MARINTEK EXPERIMENTS WITH FORM FACTOR FROM NUMERICAL CALCULATIONS
(STEEN ET AL., 2010) AND EMPIRICAL FORM FACTOR BY HOLTROP.

Model
FN RNm CTm CFm CAppm CBDm CR
0.101 3.80E+06 5.81E-03 3.58E-03 0.00E+00 5.00E-06 8.26E-04
0.121 4.56E+06 5.23E-03 3.46E-03 0.00E+00 5.00E-06 4.21E-04
0.143 5.39E+06 5.34E-03 3.36E-03 0.00E+00 5.00E-06 6.63E-04
0.157 5.91E+06 5.29E-03 3.29E-03 0.00E+00 5.00E-06 7.07E-04
0.161 6.07E+06 5.14E-03 3.28E-03 0.00E+00 5.00E-06 5.72E-04
0.181 6.83E+06 5.54E-03 3.21E-03 0.00E+00 5.00E-06 1.07E-03

Form (Steen et
Ship 0.391
factor al., 2010)
FN RNs CR CFs CVs CBDs CApps CTs
0.101 1.56E+09 8.26E-04 1.45E-03 2.17E-03 8.00E-06 0.00E+00 3.00E-03
0.121 1.87E+09 4.21E-04 1.42E-03 2.16E-03 8.00E-06 0.00E+00 2.59E-03
0.143 2.21E+09 6.63E-04 1.39E-03 2.16E-03 8.00E-06 0.00E+00 2.83E-03
0.157 2.43E+09 7.07E-04 1.38E-03 2.16E-03 8.00E-06 0.00E+00 2.88E-03
0.161 2.49E+09 5.72E-04 1.37E-03 2.16E-03 8.00E-06 0.00E+00 2.74E-03
0.181 2.80E+09 1.07E-03 1.35E-03 2.16E-03 8.00E-06 0.00E+00 3.24E-03
Form
Holtrop 0.3435
factor
0.101 1.56E+09 9.96E-04 1.45E-03 2.17E-03 8.00E-06 0.00E+00 3.17E-03
0.121 1.87E+09 5.85E-04 1.42E-03 2.16E-03 8.00E-06 0.00E+00 2.76E-03
0.143 2.21E+09 8.22E-04 1.39E-03 2.16E-03 8.00E-06 0.00E+00 2.99E-03
0.157 2.43E+09 8.63E-04 1.38E-03 2.16E-03 8.00E-06 0.00E+00 3.03E-03
0.161 2.49E+09 7.28E-04 1.37E-03 2.16E-03 8.00E-06 0.00E+00 2.90E-03
0.181 2.80E+09 1.22E-03 1.35E-03 2.16E-03 8.00E-06 0.00E+00 3.39E-03

36
5.1.1 WAVE MAKING AND RESIDUAL RESISTANCE VERIFICATION
3.00E+03
Marintek form factor
2.50E+03
Rw and Rr (kN)

2.00E+03 Steen et al. (2010) form

factor
1.50E+03
Holtrop form factor
1.00E+03
5.00E+02 KVLCC2A Rw Pressure
Integration
0.00E+00
0.09 0.14 0.19 KVLCC2A Rw Wave cut
Fn

FIGURE ‎5.1 - RW FROM NUMERICAL CALCULATIONS AND RR FROM EXPERIMENTS

In Figure ‎5.1 the results for the residual resistance from experiments and the wave resistance
from Shipflow for KVLCC2A are plotted. The wave resistance is expected to go to zero as the
Froude number approaches 0.1. The result from the wave cut approaches zero in the area of
Froude numbers between 0.14-0.16, while the wave resistance from pressure integration
approaches ~500 kN.

It can be seen from the experimental values representing the residual resistance that
components other than the wave resistance seem to dominate when the Froude number is low.
When talking to Shipflow support and Jaeouk Sun in DNV it seems to be difficult to calculate the
wave resistance at low Froude numbers. They recommended using the wave cut value, as this is
less dependent on the mesh. Somewhat lower results than the actual should be expected due to
numerical damping of waves. However, they commented that for comparison purposes it would
be the most reliable parameter.

The XPAN solver in Shipflow has been run with the finest standard mesh on the hull and free
surface. The mesh in Shipflow is optimized with regards to a standard case. After consulting with
Magnus Östberg at Shipflow support it was decided there should be no need to change the mesh
with regard to the specific case in this thesis. Thus the standard mesh was used.

2000

1500
KVLCC2A
RW (wave cut)

1000 KVLCC2B
KVLCC2C
500
KVLCC2D
0 KVLCC2E
0.1 0.12 0.14 0.16 0.18 0.2
FN

FIGURE ‎5.2 - RW FROM WAVE CUTS IN SHIPFLOW

37
 3000

RW (integrated pressure) 2500

2000 KVLCC2A
1500 KVLCC2B
1000 KVLCC2C

500 KVLCC2D

0 KVLCC2E
0.1 0.12 0.14 0.16 0.18 0.2
FN

FIGURE ‎5.3 - RW FROM INTEGRATED PRESSURE IN SHIPFLOW

In Figure ‎5.2 and Figure ‎5.3 the results from the wave making resistance calculations in Shipflow
are shown for each design, wave cuts and integrated pressure respectively. The results from the
integrated pressure method are not consistent. It seems like the sharp designs have an effect, a
hydrostatic force, which should not be there. The results for KVLCC2B and KVLCC2D both
approach zero on low Froude numbers, and seem to be describing the physics better. However,
due to the inconsistency in these results they cannot be used further in the calculations.

The results from the wave cuts seem more reasonable in relative terms between the designs.
However, they do not capture the effects that make the experimental residual results go to
around 500-700 kN at low Froude numbers in Figure ‎5.1.

The total calm water resistance will be too small when using the wave cut results, which will
influence the results of the route simulation

5.1.1.1 M ODIFICATION OF W AVE R ESISTANCE R ESULTS

The wave making resistance calculated by wave cuts in Shipflow is too low (refer to model test
results). This will influence the results as the added resistance will be relatively larger.

The non-zero residual resistance from model tests at FN 0.1 is a pressure force that should be
independent of velocity and should be equal for all the designs as the aft ships are identical. Thus
a possible modification of the results can be to add or subtract a constant value on coefficients to
get the same residual resistance coefficient for all designs at F N=0.1.

With basis in the model test results calculated with Holtrop’s form factor the target residual
coefficient at FN=0.1 is CR=9.96E-04 are found in Table ‎5.3. The difference between the calculated
results and the target value at FN=0.1 will be added to the wave resistance for each design. The
new resistance coefficient for the designs is from now on named as the modified residual
resistance. The coefficients calculated for each design can be found in ‎Appendix B.

The modified residual resistance, seen in Figure ‎5.4, becomes somewhat larger than the model
test results. However, the modified residual resistance is a better fit to the model test results than
the wave cut calculations. Thus the modified residual resistance will be used in further
calculations.

38
 2000 Marintek form factor
1800 Steen et al. (2010)
1600 form factor
RR modified (kN)

1400 Holtrop form factor

1200 KVLCC2A
1000
800 KVLCC2B
600
KVLCC2C
400
200 KVLCC2D
0.1 0.12 0.14 0.16 0.18
KVLCC2E
FN

FIGURE ‎5.4 – COMPARISON OF THE MODEL TEST RESIDUAL RESISTANCE AND MODIFIED RESIDUAL RESISTANCE

5.1.2 VISCOUS RESISTANCE VERIFICATION

The viscous resistance is calculated in the same manner by MARINTEK as explained in
chapter ‎2.1.2, with the exception of the form factor which is calculated by MARINTEK’s formula.
In Table ‎5.3 the viscous results from MARINTEK is recalculated with new form factors. These
values are used to verify that the method used is correct.

In Table ‎5.4 the viscous resistance for the model designed in Maxsurf, and used in calculations,
are shown. The vessel used in MARINTEK’s experiments should be the same, but some deviations
in wetted surface, small deviation in length in waterline, transom stern area and volume
displacement create deviations in the results.

TABLE ‎5.4 - VISCOUS RESISTANCE FOR KVLCC2A, FORM FACTOR = 0.3435

FN (1+k) (CF+ΔCF) CDB

0.1 2.09E-03 8.524E-06
0.11 2.09E-03 8.573E-06
0.12 2.09E-03 8.618E-06
0.13 2.09E-03 8.659E-06
0.142 2.09E-03 8.705E-06
0.145 2.09E-03 8.716E-06
0.155 2.09E-03 8.750E-06
0.16 2.09E-03 8.766E-06

The transom stern coefficient is somewhat larger for KVLCC2A than in MARINTEK’s results. This
is at least partially because of a smaller ratio between transom stern area and wetted surface
area, and maybe round off errors in MARINTEK’s results. However, this deviation is small, on the
order of , and is tolerated.

The viscous resistance is smaller than for MARINTEK. This deviation is also relatively small and
the method for calculating viscous resistance is assumed to be verified.

5.1.2.1 C HOICE OF M ETHOD FOR C ALCULATING F ORM FACTOR

Two methods for calculating the form factor of the designs is available. The first is Holtrop’s
formula, based on empirical values. This gives results as shown in Table ‎5.5.

39
TABLE ‎5.5 - FORM FACTORS CALCULATED BY HOLTROP'S METHOD

Design 1+k
KVLCC2A 1.343517
KVLCC2B 1.344976
KVLCC2C 1.330679
KVLCC2D 1.345783
KVLCC2E 1.311831

The fact that KVLCC2A, KVLCC2B and KVLCC2D have the largest and similar form factor seems
reasonable. It is hard to say if the decrease in form factor for KVLCC2C and KVLCC2E is realistic.

A possible alternative is to use numerical calculations to derive the form factor. KVLCC2A and
KVLCC2C were run in the XCHAP solver in Shipflow. The result was a form factor of 0.309 and
0.32 for KVLCC2A and KVLCC2C respectively. This does not seem reasonable and was a time
consuming analysis. Therefore, Holtrop’s method has been chosen and the values presented in
Table ‎5.5 are the ones used.

5.1.3 VERIFICATION OF ADDED RESISTANCE IN WAVES

The added resistance in waves is verified against results calculated in Guo (2011). Calculations
with both pressure integration and Gerritsma and Beukelmann combined with Faltinsen’s
asymptotic formula in short waves are performed and compared to results from Guo and Steen
(2011).

FIGURE ‎5.5 - EXPERIMENTAL AND NUMERICAL RESULTS FOR ADDED RESISTANCE FOR KVLCC2A (GUO, 2011)

In Figure ‎5.5 experimental and calculated results for the added resistance of KVLCC2A are shown,
taken from (Guo and Steen, 2011). “EFD” are the experimental results from the towing tank at
MARINTEK, while “EFD_Osaka” are experimental results from Osaka. Of interest is the “Com-REM”
and “Com-PI”, which is the radiated energy method and the pressure integration method
combined with Faltinsen’s asymptotic formula, respectively.

It can be seen that the pressure integration method over-predicts the added resistance in the
radiation region. The radiation energy method under-predicts in the radiation region, but not as

40
much as the over-prediction by the pressure integration method. It can also be seen that
Faltinsen’s asymptotic formula predicts well the added resistance in short waves.

The methods used by ShipX, both shown below, are the same as the methods used in Figure ‎5.5.
The radiation energy method fits best with the experimental results. Thus this will be used in the
calculations.

14
12
10
Caw = Raw/ρgζa2B2/LPP

8
6 Com-REM Heading 0 deg
4
Com-PI Heading 0 deg
2
0
0 1 2 3 4 5 6 7 8 9
Lpp/λ

FIGURE ‎5.6 - ADDED RESISTANCE COEFFICIENT FOR KVLCC2A FROM SHIPX AT FN=0.142

5.2 RESULTS
The results from calm water resistance of the ships as well as the added resistance amplitude
operator for each design and the results of the route simulation will be presented and
commented on.

5.2.1 CALM WATER RESISTANCE

Resistance coefficients calculated by the methods in chapter ‎2 are presented in ‎Appendix B.

5.2.1.1 W AVE R ESISTANCE

The wave resistance is one of the major components that are highly dependent on details in the
design. However, for low Froude numbers as the case is for KVLCC2, it is of low significance
compared to the viscous resistance. As seen in chapter ‎5.1.1, Shipflow is not able to represent the
wave resistance at these low Froude numbers accurately. In any case the relative wave resistance
between the different designs should be evaluated for comparison.

In Figure ‎5.7 the wave resistance from transverse wave cuts in Shipflow is presented for each
design. The figure only shows a small range of Froude numbers to make it possible to separate
the different designs.
RR modified (kN)

980
KVLCC2A
930 KVLCC2B
880 KVLCC2C
KVLCC2D
830
0.135 0.137 0.139 0.141 0.143 0.145 KVLCC2E
FN

FIGURE ‎5.7 - MODIFIED RESIDUAL RESISTANCE FOR A SMALL RANGE OF FROUDE NUMBERS

41
5.2.1.2 V ISCOUS R ESISTANCE
The results for viscous resistance are presented in Figure ‎5.8 below.

It is difficult to differentiate between the designs from Figure ‎5.8. The change in viscous
resistance from KVLCC2A to the other designs at FN=0.142 is shown in Table ‎5.6. The change is
calculated from;

( )
Equation ‎5-2
( )

i – index of design (B, C, D, and E)

It can be seen that the change is rather small. KVLCC2E has a 2.93% increase in resistance
compared to KVLCC2A at a given Froude number. The increase in skin friction is larger, but is
compensated by the lower form factor of KVLCC2E.

2.20E+03

2.10E+03
KVLCC2A
2.00E+03
RVS (kN)

KVLCC2B
1.90E+03 KVLCC2C
KVLCC2D
1.80E+03
KVLCC2E
1.70E+03
0.135 0.14 0.145 0.15

FN

FIGURE ‎5.8 - VISCOUS RESISTANCE FOR THE DESIGNS

KVLCC2B has a larger increase in the viscous resistance than KVLCC2C, which is mainly due to
the large increase in wetted surface.

TABLE ‎5.6 - PERCANTAGE CHANGE IN VISCOUS RESISTANCE CALCULATED IN NEWTON AT FN=0.142, WITH
KVLCC2A AS BASE CASE

Design % change
KVLCC2B 1.80 %
KVLCC2C 1.17 %
KVLCC2D 0.22 %
KVLCC2E 2.93 %

42
5.2.1.3 T OTAL CALM W ATER R ESISTANCE
The total calm water resistance is shown in Figure ‎5.9 for a small range of Froude numbers.
These results will be the input to the route simulation in ShipX.

3300

3100 KVLCC2A modified

RT (kN)

KVLCC2B modified
2900
KVLCC2C modified
2700
KVLCC2D modified
2500
KVLCC2E modified
0.135 0.14 0.145 0.15
FN

FIGURE ‎5.9 – TOTAL CALM WATER RESISTANCE

KVLCC2E has the highest calm water resistance against Froude number. The total viscous
resistance coefficient is lower for the longer designs due to a longer water line length would that
result in a higher Reynolds number at constant speed. Remembering the ITTC curve for friction a
higher Reynolds number results in a lower plate friction coefficient for a given speed. Thus some
of the increased wetted surface will be compensated for by a lower friction coefficient as well as a
lower form factor. The resistance curves plotted against Froude number will for the longer
designs, KVLCC2C and KVLCC2E, will shift to the right when plotted against speed because these
designs will have a higher speed corresponding to a Froude number. For a given Froude number
one can find the relation between the speeds for the short designs and longer designs.

√ Equation ‎5-3

This ratio is 1.0122 for KVLCC2C and 1.0243 for KVLCC2E. Figure ‎5.10 shows the total resistance
plotted against velocity instead of Froude number. Thus the expected increase in calm water
resistance for the sharper designs is not the case by these calculations. Due to the relatively low
increase in wetted surface for KVLCC2C it is not punished as much as expected. KVLCC2E also has
a low calm water resistance but it seems like the benefit of the length is punished by a larger
increase in the wetted surface.

3300
3200
3100
3000 KVLCC2A modified
RT (kN)

2900 KVLCC2B modified

2800 KVLCC2C modified
2700
KVLCC2D modified
2600
2500 KVLCC2E modified
14.8 15.3 15.8 16.3
V (kn)

FIGURE ‎5.10 - TOTAL CALM WATER RESISTANCE PLOTTED AGAINST KNOTS

43
5.2.2 ADDED RESISTANCE IN WAVES
The added resistance in waves for headings 0°, 45°, 90°, 135° and 180° is shown in Figure ‎5.11. In
the radiation region 45° heading has the largest added resistance. However, it has lower added
resistance in the diffraction region than head sea waves. Thus what is the most critical heading,
in terms of added resistance, is dependent on the wave conditions. The same tendency is seen for
all the other designs as well, but not presented here. Plots for all headings at FN=0.142 is given for
all designs in ‎8Appendix C.

In Figure ‎5.12 the added resistance for each design at FN=0.142 in head sea is shown. It can be
seen that KVLCC2C and KVLCC2E has a slightly higher added resistance coefficient, C aw, in the
diffraction region. However, it is pronouncedly lower in the diffraction region, as expected. The
difference between KVLCC2C and KVLCC2E is, however, not as significant.

8.00E+00
6.00E+00
4.00E+00 0 degrees
2.00E+00 45 degrees
Caw

0.00E+00 90 degrees
0 5 10 15 135 degrees
-2.00E+00
180 degrees
-4.00E+00
-6.00E+00
Lpp/λ

FIGURE ‎5.11 - RAO IN ADDED RESISTANCE FOR KVLCC2A AT DIFFERENT HEADINGS, F N=0.142

8
7
6
5 KVLCC2A
4 KVLCC2B
3 KVLCC2C
Caw

2 KVLCC2D
1
KVLCC2E
0
0 2 4 6 8 10
Lpp/λ

FIGURE ‎5.12 - ADDED RESISTANCE RAO AT HEAD SEAS AT FN = 0.142

It can be seen that KVLCC2A has more similar results compared to KVLCC2D than KVLCC2B.
KVLCC2B should have the same waterline curve as KVLCC2A and they were thus expected to
have the same results in the diffraction region. One explanation might be that the bulb of
KVLCC2A influences the calculations even though it is below the design water line in calm water.
Another possibility is that the ShipX model, which was the same used by Bingjie Guo in (Guo,
2011), is slightly different from KVLCC2A waterline, which had to be redesigned in Maxsurf. A
test with the model used to calculate the calm water resistance has been done. This design has

44
the same water line curve as KVLCC2A. The results comparing KVLCC2A, KVLCC2B (from Guo)
and KVLCC2B (designed in Maxsurf without line drawings) are shown in Figure ‎5.13.

It should be emphasized that the redesign was done without any line drawings, and thus the
results are not expected to be exactly the same for the two models. However, one can see that the
redesigned model has even lower added resistance than KVLCC2B from (Guo, 2011). The
redesigned model has the same water line curve as KVLCC2A, which thus indicates that the bulb
of KVLCC2A influences the results even at short waves.

8
7
6
5 KVLCC2B (Guo, 2011)
Caw

4
KVLCC2B (Redesign
3
Maxsurf)
2 KVLCC2A
1
0
0 5 10 15
Lpp/λ

FIGURE ‎5.13 - COMPARISON OF ADDED RESISTANCE CALCULATIONS IN SHIPX FOR KVLCC2B MODELS, FN=0.142

5.2.3 SIMULATION OF ROUTES

With the scatter diagrams, calm water resistance and added resistance in waves established the
propeller and engine characteristics are the last properties to establish before evaluating each
design over the routes.

5.2.3.1 P ROPELLER AND E NGINE C HARACTERISTICS

For ShipX to calculate the speed-loss the propeller and engine characteristics have to be
established. ShipX calculates a corrected open water diagram, loss of propeller efficiency, due to
variable loading in waves.

The propeller characteristics are given for experiments to standardize the experiments
performed all over the world for CFD validation (Simman2008, 2008). The given propeller
characteristics have been used on a Wageningen B-series propeller, a widely used propeller
series, which is included in the ShipX library.

From Sea-web (Sea-Web, 2012), statistics for the fleet of ships can be found. A search on engine
and propulsion configuration for tankers and bulk carriers between 290 and 330 meters long
gives a basis for statistically determine the most used configurations. It shows that ~92% of the
vessels fulfilling the criteria of the search have 1 propeller, and even more have fixed pitch as
well.

The propeller characteristics given by Simman2008 (2008) are shown in Table ‎5.7.

45
TABLE ‎5.7 - PROPELLER CHARACTERISTICS

Type Fixed
No. of blades 4
D (m) 9.86
P/D (0.7R) 0.721
Ae/A0 (m) 0.431
Rotation Right hand
Hub ratio 0.155

ShipX also needs input on the propulsion characteristics. That is the wake coefficient, , thrust
deduction coefficient, t, rotational efficiency, , mechanical efficiency, , and gear box
reduction ratio. The first coefficients are taken from Kim et al. (2001), the mechanical efficiency
is assumed to be 0.97 and the gearbox reduction ratio is assumed to be one, as the comparison
ships from Sea-Web had a majority of direct drive propulsion systems. Values are presented in
Table ‎5.8. They are assumed equal for all the designs as the aft hull has major influence on the
values and are equal for all the designs.

TABLE ‎5.8 - PROPULSION SETTINGS

RPM 80
w 0.305
t 0.19
1.005
0.97

A typical engine size for a ship of similar dimensions as KVLCC2 can be found by using the same
Sea-web search. A plot of the length and engine power for tankers and bulk carriers is shown in
Figure ‎5.14. It can be seen that the engine size is between 22 000 kW and 27 000 kW. Since the
KVLCC2 designs are in the upper range of the vessels here 27 000 kW is chosen as engine size in
the analyses in ShipX.

28000
Engine size (kW)

26000

24000

22000 Comparison Ships

20000
315 316 317 318 319 320
Length (m)

FIGURE ‎5.14 - ENGINE SIZE OF COMPARISON SHIPS FROM SEA-WEB.

5.2.3.2 S PEED -L OSS R ESULTS

The speed-loss calculations are calculated for each design over each route. As mentioned the
scatter diagrams do not contain any information on wave direction. Thus the results are first
presented as speed-loss over headings 0°-180° with 45° intervals. Average speed-loss results,
where each heading is weighted equally, are also presented.

Due to different calm water resistance, each ship has a different attainable speed in calm water
presented in Table ‎5.9. The table shows that both KVLCC2C and KVLCC2D have a higher

46
attainable speed for the engine and propulsion configuration presented above in calm water than
the blunter designs.

TABLE ‎5.9 - ATTAINABLE SPEED IN CALM WATER BASED ON ENGINE AND PROPULSION CONFIGURATION
PRESENTED IN CHAPTER ‎5.2.3.1.

Design VattCW (kn)

KVLCC2A 13.4
KVLCC2B 13.26
KVLCC2C 13.53
KVLCC2D 13.33
KVLCC2E 13.44

Speed-loss on each route for each design over headings is presented in Figure ‎5.15-Figure ‎5.19.

0.5
Speed change (kn)

0 AG-LOOP
-0.5 0 45 90 135 180
AG-Chiba
-1
Mongstad-NY
-1.5
Tubarao-Quindao
-2
Heading (deg)

FIGURE ‎5.15 - KVLCC2A: SPEED-LOSS OVER HEADINGS FOR EACH ROUTE

0 AG-LOOP
Speed change (kn)

0 50 100 150 200

AG-Chiba
-1
Mongstad-NY
Tubarao-Quingdao
-2
Heading (deg)

FIGURE ‎5.16 - KVLCC2B: SPEED-LOSS OVER HEADINGS FOR EACH ROUTE

0.5
Speed change (kn)

0 AG-LOOP
-0.5 0 45 90 135 180
AG-Chiba
-1
Mongstad-NY
-1.5
Tubarao-Quindao
-2
Heading (deg)

FIGURE ‎5.17 - KVLCC2C: SPEED-LOSS OVER HEADINGS FOR EACH ROUTE

47
 0.5

Speed change (kn)

0 AG-LOOP
-0.5 0 45 90 135 180
AG-Chiba
-1
Mongstad-NY
-1.5
Tubarao-Quindao
-2
Heading (deg)

FIGURE ‎5.18 - KVLCC2D: SPEED-LOSS OVER HEADINGS FOR EACH ROUTE

0.5
Speed change (kn)

0 AG-LOOP
-0.5 0 45 90 135 180
AG-Chiba
-1
Mongstad-NY
-1.5
Tubarao-Quindao
-2
Heading (deg)

FIGURE ‎5.19 - KVLCC2E: SPEED-LOSS OVER HEADING FOR EACH ROUTE

For comparison purposes the speed-loss over each heading for all the designs on the route
Arabian Gulf to LOOP has been plotted in Figure ‎5.20. The other routes have the same trends
seen in the figures above.

0.5
Speed change (kn)

KVLCC2A
0 KVLCC2B
0 45 90 135 180
KVLCC2C
-0.5
KVLCC2D
-1 KVLCC2E
Heading (deg)

FIGURE ‎5.20 – SPEED-LOSS FOR EACH DESIGN ON THE ROUTE FROM THE ARABIAN GULF TO LOOP

As expected the route from Mongstad to New York gives the largest effect on the speed-loss
calculations. The benefit of the sharper bow is also, as expected, greatest in head seas.
Comparison of these two designs to the three other designs shows that at 45° heading the
difference in speed loss is not very significant.

At 90° heading the speed-loss is actually larger for the sharpest designs. This might be due to the
longer bow creating a larger wall surface for the waves to reflect off. However, the difference is
very small and on larger headings the effect of a sharper bow can be neglected. This is no
surprise as the waves will not face the bow part of the vessel.

Figure ‎5.21 and Figure ‎5.22 show the speed-loss averaged over headings for each design on each
route. The weighting factor is assumed to be one for each heading, i.e. the vessel will experience
sea states equally from all sides.

48
 -0.05
Speed change (kn) KVLCC2A
-0.15 KVLCC2B
KVLCC2C
-0.25
KVLCC2D
-0.35 KVLCC2E

-0.45

FIGURE ‎5.21 – SPEED-LOSS AVERAGED OVER HEADINGS GROUPED BY ROUTE

0
KVLCC2A KVLCC2B KVLCC2C KVLCC2D KVLCC2E
Speed change (kn)

-0.1 AG-LOOP
AG-Chiba
-0.2
Mongstad-NY
-0.3
Tubarãro-Quindao
-0.4 Voyage average
-0.5

FIGURE ‎5.22 - SPEED-LOSS AVERAGED OVER HEADINGS GROUPED BY DESIGN

The voyage average shown in Figure ‎5.21 and Figure ‎5.22 is the average speed loss over the
routes, as if a vessel were to trade the same 4 routes in cycle, weighted by the route length. The
short route of Mongstad to New York compared to the routes departing from the Arabian Gulf
has little influence on the overall average.

To investigate if there is a difference in the benefit of a sharper bow on the different routes,
representing a range in typical sea states the speed-loss of KVLCC2C has been compared to the
speed-loss of KVLCC2A. This shows that the speed loss of KVLCC2C is 14.2 % lower on the route
from the Arabian Gulf to LOOP, 13.8% from the Arabian Gulf to Chiba and 16.2% lower on the
route from Mongstad to New York across the North Atlantic. The route from Tubarão to Qingdao
shows a 14.9% lower speed-loss for KVLCC2C.

13.3
Attainable speed in waves

13.2
13.1
13
12.9 KVLCC2A
12.8
(kn)

12.7 KVLCC2B
12.6 KVLCC2C
12.5
KVLCC2D
KVLCC2E

FIGURE ‎5.23 - ATTAINABLE SPEED IN WAVES AVERAGED OVER HEADINGS GROUPED BY ROUTE

49
 13.4

Attainable speed in waves

13.3
13.2
AG-LOOP
13.1
13 AG-Chiba
(kn)

12.9 Mongstad-NY
12.8 Tubarão-Quingdao
12.7
Voyage average
12.6
12.5
KVLCC2A KVLCC2B KVLCC2C KVLCC2D KVLCC2E

FIGURE ‎5.24 - ATTAINABLE SPEED IN WAVES AVERAGED OVER HEADINGS GROUPED BY DESIGN

Figure ‎5.23 and Figure ‎5.24 shows the attainable speed for each design with added resistance
and the open water diagram corrected for waves included. It should be noted that both KVLCC2C
and KVLCC2E have a higher attainable speed in calm water. However, to see the effect of the
added resistance on each design the percentage speed-loss of the attainable speed in calm water
for each design is presented in Table ‎5.10.

KVLCC2C and KVLCC2E have more or less the same results. The slightly lower head sea added
resistance of KVLCC2E in the diffraction regime is not enough to make KVLCC2E significantly
better than KVLCC2C. The blunter designs and KVLCC2D especially are the one affected the most
by waves, as expected. However, KVLCC2B has overall lowest performance due to both a high
calm water resistance and a large influence from waves.

TABLE ‎5.10 - PERCENTAGE SPEED-LOSS OF ATTAINABLE SPEED IN CALM WATER

Voyage
AG-LOOP AG-Chiba Mongstad-NY Tubarão-Qingdao
average
KVLCC2A 2.22 % 2.03 % 3.12 % 2.39 % 2.33 %
KVLCC2B 2.35 % 2.16 % 3.32 % 2.54 % 2.47 %
KVLCC2C 1.88 % 1.73 % 2.59 % 2.02 % 1.97 %
KVLCC2D 2.37 % 2.20 % 3.33 % 2.56 % 2.50 %
KVLCC2E 1.86 % 1.70 % 2.58 % 2.01 % 1.95 %

50
6 CONCLUSION
From the calculations performed in this thesis it can be concluded that

 KVLCC2C is the overall best performer when both calm water resistance and influence of
the wave environment is included. Even though the water line length of the design has
been increased by 8 m, the calm water resistance is relatively low and the benefit of the
sharper bow also contributes to better performance in waves than the blunter designs.
 KVLCC2E has the second highest attainable speed. It has smaller added resistance in the
diffraction regime than all the other designs. However, the benefit of this is decreased by
a higher calm water resistance than KVLCC2C thus giving a lower attainable speed. A
design like KVLCC2E, which is extended further than the overall length of the original
KVLCC2 design, may not be desired from an owner’s point of view because it may violate
length restrictions in ports and thus reduce the number of possible ports to visit.
 Of the designs presented KVLCC2B has the lowest attainable speed on each route. This is
not a surprise as the vertical sides all the way down to the bottom create a large viscous
resistance. It was expected that the speed-loss of KVLCC2B was going to be similar to
KVLCC2A due to the equal water line curve of the bow. However, the results show that
the speed-loss of KVLCC2B is very similar to KVLCC2D. This is despite the results from
added resistance calculations showing that KVLCC2B design has a lower added
resistance coefficient in the diffraction regime than KVLCC2A.
 KVLCC2D has the lowest attainable speed of the designs with a naturally shaped under
water hull, excluding KVLCC2B. The speed loss is not substantially much greater than for
KVLCC2A. However, a larger calm water resistance contributes to a poor overall
performance.
 The fact that the effect is greatest for head seas shows that a sharper bow will reduce the
added resistance in waves for head seas, which is the primary goal of the sharper
entrance angle. Also the sharper entrance angle does not influence the performance in
the radiation regime or in oblique waves too much. It also shows that the reflection effect
on added resistance, despite the low magnitude of the effect compared to radiation, is a
major component of the total added resistance. Thus the majority of wave conditions
experienced are small (short) compared to KVLCC2 design.
 The dominating wave heading on a route seems to play a crucial role when evaluating
different designs. The benefit of a sharper bow will be smaller if the vessel travels in
oblique waves most of the time. However, a merchant vessel will seldom trade the same
cycle its entire operating time. Thus an assumption of equal importance of all headings is
not that far-fetched.

Following the argumentation of headings above the results indicate that a sharpening of
the bow would be beneficial even though the gain of a sharper bow is mostly within
wave headings of 45°.
 A small increase in benefit is seen on routes with more severe wave conditions as in the
North Atlantic. However, it does not influence enough to give a different optimal design.

The calm water resistance did not turn out as was expected in chapter ‎3.3.8. The conclusions
stated above have to be based on the results calculated in this thesis. However, as discussed in
the previous chapter several aspects of the calm water resistance have to be assessed in more
detail and to a higher precision to be able to confidently state the benefits of a sharper bow. As it
turns out, the difference between the designs is not extreme, and thus the precision requirement
for a calm water resistance analysis is high.

51
The wave making resistance calculation is one of the factors that should be calculated with other
methods as Shipflow was not able to give reasonable results. The form factor calculated by
Holtrop’s empirical method is also an uncertain parameter in the calculations. A precise division
of the calm water resistance between the designs is dependent on these two parameters.

52
7 INNOVATIVE BOW DESIGNS
There are several bow designs developed with the objective of reducing the added resistance in
waves. In this thesis they are referred to as innovative bow designs, as they have features not
found on most bow designs today with special consideration on added resistance in waves. The
last decade some segments have opened up to unconventional bows, and some designs have
become a relatively common feature in the fleet. This is probably most noticeable in the offshore
service segments, where X-Bow from Ulstein Design and the STX OSV’s relatively novel design
has become the standard bow design on ships built at Ulstein Shipyard and STX’ yards in Norway.
Designs with these bows are also sold to yards and ship owners all over the world.

In other segments and especially the larger segments of merchant ships the trend of highly
innovative bow has not been the same. The offshore service fleet is generally small compared to
the merchant ships supplying the world trade. Still some designs have been developed especially
for large segments, and at least one has been built and tested in full scale.

This chapter will review some innovative bows and discuss their working principles and evaluate
their applicability to large, slow ships. Most bow designs are hard to review as they are subject to
patent protection and information from experiments and such are often confidential. The
efficiency of such bows can only be concluded by model tests, CFD calculations or full scale tests.
However, an evaluation of the working principles and how well they fulfil the objectives
according to theory will be carried out.

7.1 X-BOW

FIGURE ‎7.1 - THE X BOW CONCEPT

It is not easy to acquire experimental data on this bow. However, the patent application provides
some information on the reasoning for developing this bow, and the effects that are aimed for by
the bow design. The website of Ulstein Group provides some information, however, rather
superficially.

7.1.1 BACKGROUND
The objective of the X bow design is not merely to reduce added resistance in waves. It aims at
improving several aspects of the operation of offshore vessels. Offshore service vessels often

53
have the wheelhouse and superstructure in front of the mid ship. Thus motions are a critical
point in this case for the accommodation and resting of the crew. Motion characteristics are also
an important aspect of the operability of offshore vessels, which Ulstein Design claims to have
improved with X bow. The patent application sums up advantages of X bow as (amongst
others)(Kvamsvåg, 2006):

- Lower accelerations and retardations, which give higher average speed at sea, thereby
reducing power requirement and consumption of fuel.
- Reduction in the amount of or elimination of green water on deck.
- Lower risk of heavy weather damage to the foreship because the reflection of waves is
reduced.
- Improved working environment on board with regard to;
o The above mentioned lower accelerations and retardations, improving safety
during navigation and higher operability.
o Reduction of noise and vibration because of gentle motions and reduced
slamming. Increasing comfort and better environment with regard to resting
time of the crew.

Even though the patent application states validity in general for displacement ships, it focuses on
vessels intended for marine operations. It states that when offshore service vessel are waiting on
the weather, they will normally have their bow turned upwind and into the waves. Typical bows
on offshore ships have raked bow above the waterline resulting in reflection of waves to a great
degree, heavily subject to slamming of waves against the ship side, generation of spray and
occurrence of green water on deck in front of the superstructure, (Kvamsvåg, 2006).

7.1.2 WORKING PRINCIPLES

The design is briefly described as a backward sloping bow above the water line as seen in
Figure ‎7.1. The underwater hull is similar to conventional hulls, thus the concept focus on the hull
shape above the waterline.

Volume has been moved from the shoulders and the common flare forward and up resulting in
smoother volume distribution in the bow. The conventional flare is designed mainly to reduce
green water on deck. With the X-bow, the waves are allowed to creep up on the bow and thus the
tall enclosed bow structure of the design is needed.

FIGURE ‎7.2 - WATERLINE OF X-BOW, (KVAMSVÅG, 2006).

The relocation of volume from the shoulders and forward allows for a sharper entrance angle
and thus less reflection of waves. The waterline is shown in Figure ‎7.2.

54
For conventional bows, when encountering incident waves, the submerged volume increases
rapidly, buoyancy increases correspondingly, which results in the waves being thrown back
causing large accelerations and retardation forces. The smoother lines of the X bow above water
allow the wave to follow the ship side and be directed to the sides, resulting in smoother motions
and less throwback of water in large waves.

7.2 STX DESIGN

FIGURE ‎7.3 - STX OSV - PSV 08 (HTTP://WWW.STXOSV.COM/NEWSANDMEDIA/PAGES/DEFAULT.ASPX)

7.2.1 BACKGROUND
STX OSV designs and builds offshore and specialized vessels used in the offshore oil and gas
exploration and production and oil services industries. They have also developed an innovative
bow concept, obviously focused on the segments mentioned above. This bow design is
proprietary information and therefore, test results are not available. The patent application,
(Tvete and Borgen, 2012), has been used as the main source of information.

On the same basis as for the X-bow, the bow is developed to improve performance in marine
operations. The patent application states that the objective of the new bow is to reduce the added
resistance in waves, as well as reducing the level of accelerations, motions in seaway and
slamming in the bow region. This should result in less fuel consumption and more comfort for the
crew, (Tvete and Borgen, 2012).

7.2.2 WORKING PRINCIPLES

The bow concept is divided into three sections, dealing with three different aspects of the
resistance regime.

On the lower part, section A in Figure ‎7.4, the hull shape is kept more or less as conventional
hulls to minimize the calm water resistance. This part contributes to a lesser degree to the added
resistance in waves.

The middle part, B, comprises a blunt shaped surface curving inwards and backwards from the
transition area and back to a recessed blunt shaped section, designed to reflect smaller waves to
the sides. This is as opposed to having a very sharp curve in this area which, according to Tvete
and Borgen (2012), will make the small waves follow the hull shape further aft and induce

55
friction resistance. It is argued that the short time period of handling the reflection of waves will
generate less speed-loss than the friction resistance of following waves.

FIGURE ‎7.4 - PROFILE VIEW OF THE STX BOW, (TVETE AND BORGEN, 2012).

The upper bow portion, C, is stretched forward making it sharper. This makes it possible to
reduce the flare angles, seen on conventional bows, and shoulder of the bow obtaining some of
the same characteristics as the X bow with slower volume increase when piercing into the large
waves. The sharpening is based on the same reasoning as the X-bow, making the bow pierce
through the waves and split them instead of hitting the waves with a blunt bow design.

7.3 AX-BOW
7.3.1 BACKGROUND
The Ax-bow is as bow concept designed by NKK. NKK is a Japanese steel manufacturer and ship
builder. The Ax-bow concept was developed because the energy saving measures in later years
have led to a smaller power supply on merchant ships, thus the speed-loss of these energy-saving
ships is crucial compared to ships with a large sea margin implemented. Hence, a ship with better
performance in waves even with smaller installed propulsion power was desired, (Matsumoto,
2002).

The bow concept was installed on “Kohyohsan”, a 172 000DWT Cape size bulk carrier with
overall length of 289 m and breadth 45 m. A sister ship was fitted with an ordinary bow and full
scale measurements have been done.

56
 FIGURE ‎7.5 - THE AX-BOW CONCEPT COMPARED TO THE ORDINARY BOW AND LEADGE-BOW (ITTC, 2008).

7.3.2 WORKING PRINCIPLES

The design focuses on the shape above the waterline. The hull shape below the waterline is the
same as for the ordinary bow maintaining the calm water performance of the ordinary ship. The
bow has been sharpened above the waterline, like a beak, to reflect incident waves more to the
sides than forward. To satisfy the length requirements in ports the bow had to be sharpened as
much as possible within the regulations. A requirement of 289 m maximum overall length in a
European port was the dimensioning requirement, (Hirota et al., 2005).

Model tests have been performed with the bow. Figure ‎7.6 shows the results as the non-
dimensional resistance increase in regular waves. The figure indicates a 20 to 30 % reduction in
almost the entire range of wavelength, (Matsumoto, 2002). However, the reduction seems to be
largest in the small (short) waves where a reduction of almost 50 % is measured.

Full scale measurements are summed up in Hirota et al. (2005), see Figure ‎7.8. They show that
the Ax-bow reduces the speed-loss, especially in the higher sea states compared to a
conventional ship. Values are given in Table ‎7.1. From track records of the Kohyohsan it is shown
that the required horse power is reduced by 3-4 % compared to the sister ship (Sea-Japan, 2006).

57
 TABLE ‎7.1 - REDUCTION RATIOS OF SPEED-LOSS, (HIROTA ET AL., 2005).

FIGURE ‎7.6 - ADDED RESISTANCE RESULTS FROM MODEL TESTS WITH AX-BOW

7.4 BEAK BOW

The Beak bow is a predecessor of the Ax-bow. It is designed as a longer and pointy beak, seen in
Figure ‎7.7. Due to the mentioned length restrictions the Ax-bow was developed. The Beak bow
tested in Hirota et al. (2005) would give a ship with overall length of 300 m.

There is not much difference in the measured added resistance coefficient between the Beak bow
and the Ax bow, hence approximately 20-30% reduction in resistance increase in waves
compared to conventional bows.

58
 FIGURE ‎7.7 - BEAK BOW PROFILE AND WATERPLANE VIEW, (Hirota et al., 2005).

7.5 LEADGE BOW

The Leadge bow is a further development of the Ax-bow, shown in Figure ‎7.5. The working
principles of the bow are the same as the Ax-bow, reflecting waves to the sides. However, the
stem has been straightened filling up the gap between the “Ax” and the bulb. This eliminates the
bulbous bow in the profile view. The whole stem line is sharpened, also under the waterline hull.
Due to this the bow was expected to reduce the added resistance in both ballast and full load
conditions. The design approach of the Leadge bow is very similar to that of KVLCC2C and
KVLCC2E, thus a confirmation on the performance regarding added resistance can be made.

According to Sea-Japan (2006) the bow shape has been confirmed to give the same calm water
wave making resistance as the conventional hull with bulbous bow. Due to the relative low speed
of the vessel, the wave making resistance is a small part of the total resistance and a change
would not contribute much to the overall calm water resistance. Thus an increase in wave
making resistance is not crucial to the overall performance, (Hirota et al., 2005).

A curious result is given in (Hirota et al., 2005). It states that Ax-bow has a reduction of 20-30%
in one part of the paper. However, in another part treating the Leadge bow, it states that the
reduction is 12% in head seas for the Ax bow, and the Leadge bow has a reduction of 19%. Two
different models have been used in the cases and the ordinary bow seems to be different in the
efficiency regarding added resistance. In any case, the latter results seem to point to the
increased efficiency of the Leadge bow regarding added resistance in waves.

59
7.6 EVALUATION OF DESIGNS IMPLEMENTED ON KVLCC2
The X-bow and STX’s bow are designed with the offshore service segments in mind. These ships
are typically 70-140 m long. Thus, the amount of waves in the short wave region is considerably
lower than for a VLCC.

The conventional bows in the offshore segments are also considerably less blunt than the large
tanker and bulk segments, probably due to a higher Froude number. The wave making resistance
is more significant than for large slow ships, and a sharper bow may be beneficial for the calm
water resistance. Thus the gain of sharpening the bow of these ships is expected to be less for
two reasons. The conventional bow is considerably sharper and the amount of waves that can be
included in the short wave region is considerably less than for large tankers.

The argumentation from STX that if the shape of the mid-section is too sharp the waves will not
be reflected off the ship, but rather guided along the ship side seems reasonable. However, the
significance of the friction resistance compared to the reflection in this case is questionable and
difficult to evaluate without any test results. Without much information it is also assumed that
the bluntness of the section is not in the range of tankers and bulk carriers, rather a relative term
considering conventional bows on offshore service vessels. Thus, it is probably sharper than the
average VLCC bow.

The exposure time to waves inducing added resistance through heaving and pitching is larger
than for ships like KVLCC2, due to the significant difference in size. Thus the X-bow and STX’ bow
have a great focus on this region. The pointy shaped section of the bows are primarily to split the
large waves, instead of throwing them back, which requires a lot of energy and creates speed-loss.

The X-bow allows waves to creep up on the bow as it slopes backwards. This requires a relatively
tall bow. On the X-bow and for STX it is a benefit as the superstructure can be incorporated in this
tall bow. On a VLCC this would probably require a taller bow than customary today due to issues
with green water on deck. The implementation of this may not be beneficial as more wind is
captured by the bow.

On the other hand, a sharper bow on a VLCC would work in a similar way, splitting the waves
reducing the throwback caused by the rake and flare on a conventional bow. Thus the large
waves may cause issues with green water on deck, and a taller bow may be necessary.

FIGURE ‎7.8 - FULL SCALE MEASUREMENTS OF AX BOW FITTED ON A CAPE-SIZE BULK CARRIER, (Hirota et al.,
2005)

For large, slow ships the exposure time to the range of relatively short waves is larger and thus
the reflection of waves is more important. As seen in the full scale measurements of the Ax-bow

60
in Figure ‎7.8 the trending benefit of the Ax-bow is increasing with wave height. Thus it seems like
the sharpening above the waterline also influences the added resistance in the radiation region
as well as in the reflection region.

Universal Shipbuilding Corp. applied for the patent on the Leadge-Bow in 2002, (Sea-Japan,
2006). However, Financial Times states that the patent was denied on the basis that it looks too
similar to old-fashioned, low-tech hulls, (Dickie, 2011). This shows that improvement in
performance in seaway can be improved by simple, not necessarily drastically innovative means.

61
8 FURTHER WORK
As discussed in the results and conclusion several aspects of the calm water analysis needs to be
taken a closer look at. A method for calculating the wave making resistance of each design
precisely, with effects like wave breaking, should be found. A possibility is to use RANS
calculations with VOF. Another possibility would be to run model tests on several designs. Both
methods could give a resistance curve more precisely than the XPAN solver in Shipflow. The
methods could also provide form factors that are based on the actual hull shape and not on
empirical formulas. However full scale CFD calculations are extremely time consuming and
access to a cluster or similar is preferable.

The last time dependent objective of the problem definition for this thesis which deals with a
numerical study of innovative bow shapes is a natural extension of this thesis. Effort should be
made on searching for even more developed innovative bow designs for large, slow ships to be
able to compare different approaches to reduce the added resistance. Possibly developing a new
design from scratch could be an interesting, but time consuming task. CFD will have to be utilized
to evaluate the efficiency of designs without vertical sides. This is time consuming and another
issue is to get access to patented designs to compare with.

Several simplifications and assumptions have been made in this thesis. One expansion of the
problem could be to include an operational profile and direction distribution of the wave
environment to get an even more realistic evaluation of the performance of each design.

Another could be to include more detailed engine characteristics such that fuel consumption can
be calculated and a study of how different bow designs influence the economics. This diverges
from a hydrodynamic study, but is none the less interesting.

62
 References

ARRIBAS, F. P. 2007. Some Methods to Obtain the Added Resistance of a Ship Advancing in Waves.
Ocean Engineering, 34, 946-955.
AZCUETA, R. Steady and Unsteady RANS Simulations for Littoral Combat Ship. Proceedings of
the 24th Symposium on Naval Hydrodynamics, 2004 St. Johns, Newfoundland and
Labrador, Canada.
BOESE, P. 1970. Simplified Method of Calculation of Increased Ship Resistance in Waves (Eine
einfache methode zur berechnnung der winderstansershohung eines schiffes im
seegang). Shiffstechnik, 17 (86).
CHINA-DAILY. 2010. China's iron-ore imports from South Africa, Ukraine, Canada up in 2009
[Online]. China Daily. Available: http://www.chinadaily.com.cn/china/2010-
02/16/content_9471501.htm [Accessed 22.03 2012].
CHOI, J. & YOON, S. B. 2009. Numerical Simulations Using Momentum Source Wave-maker
applied to RANS Equation Model. Coastal Engineering, 56, 1043-1060.
DICKIE, M. 2011. Japanese Yards Claim to Have Edge Over Rivals. Financial Times Tokyo Financial
Times.
DNV 2010. Triality - Taking Environmental and Economical Performance a Long Step Forward.
Tanker Update, DNV. DNV.
FALTINSEN, O. M. 1990. Sea Loads On Ships And Offshore Structures, Cambridge, Cambridge
University Press.
FALTINSEN, O. M. 2010. Hydrodynamics of High-Speed Marine Vehicles, Cambridge New York,
Cambridge University Press.
FALTINSEN, O. M., MINSAAS, K. J., LIAPIS, N. & SKJØRDAL, S. O. Prediction of Resistance and
Propulsion of a Ship in a Seaway. Proceeding of the 13th Symposium on Naval
Hydrodynamics, 1980. 505-529.
FALTINSEN, O. M. & SVENSEN, T. Incorporation of Seakeeping Theories in CAD. In:
OORTMERSSEN, G. V., ed. Proceedings of the International Symposium on CFD and CAD
in ship design, 1990 Wageningen, the Netherlands. Elsevier Science Publishers B.V.
FATHI, D. & HOFF, J. R. 2010. ShipX Vessel Responses (VERES) - Theory Manual. Trondheim:
MARINTEK A/S.
FUJII, H. & TAKAHASHI, T. Experimental Study on the Resistance Increase of a Ship in Regular
Oblique waves. Proceedings of the 14th ITTC, 1975 Ottawa, Canada. 351-360.
GENCO. 2012. Industry [Online]. Available: http://www.gencoshipping.com/industry.html
[Accessed 22.03 2012].
GERRITSMA, J. & BEUKELMAN, W. 1972. Analysis of the Resistance Increase in Waves of a Fast
Cargo Ship. International Shipbuilding Progress, 19, 285-292.
GUO, B. 2011. Numerical and Experimental Investigation of Added Resistance in Waves. PhD,
Norwegian University of Science and Technology.
GUO, B. & STEEN, S. Added Resistance of a VLCC in Short Waves. Proceedings of the ASME 2010
29th International Conference on Ocean, Offshore and Arctic Engineering, 2010a
Shanghai, China.
GUO, B. & STEEN, S. 2010b. Experiment on Added Resistance in Short Waves. 28th Symposium on
Naval Hydrodynamics. Pasadena, CA, USA.
GUO, B. & STEEN, S. 2011. Comparative Study of Sea-Keeping Prediction of Ships in Head Sea
Waves. Trondheim: Norwegian University of Science and Technology.
GUO, B., STEEN, S. & DENG, G. B. 2011. Sea-keeping Prediction of KVLCC2 in Head Waves with
RANS. Doctoral Thesis: Numerical and Experimental Investigation of Added Resistance in
Waves, 126-139.
HAVELOCK, T. H. 1942. Drifting Force on a Ship among Waves. Philosophical Magazine.
HIROTA, K., MATSUMOTO, K., TAKAGISHI, K., YAMASAKI, K., ORIHARA, H. & YOSHIDA, H.
Development of Bow Shape to Reduce the Added Resistance due to Waves and
Verification on Full Scale Measurement. International Conference on Marine Research
and Transportation, 2005 The Island of ISCHIA (Gulf of Naples) Italy.
ITTC, S. C. Final Report and Recommendation to the 25th ITTC. Proceedings of the 25th ITTC. The
Seakeeping Committee. 2008 Fukuoka City.

63
JOOSEN, W. P. A. Added resistance in waves. Proceeding of the Sixth Symposium on Naval
Hydrodynamics, 1966 Washington.
KASHIWAGI, M. Impact of Hull Design on Added Resistance in Waves - Application of the
Enhanced Unified Theory. Proceedings of the 10th International Marine Design
Conference, 2009 Trondheim, Norway. 521-535.
KASHIWAGI, M., IKEDA, T. & SASAKAWA, T. 2010. Effect of forward speed of a ship on added
resistance in waves. International Journal of Offshore and Polar Engineering, 20, 196-203.
KIM, W. J., VAN, S. H. & KIM, D. H. 2001. Measurement of flows around modern commercial ship
models. Experiments in Fluids, 31, 567-578.
KVAMSVÅG, U.-D.-A. Ø. G. 2006. The Foreship Arrangement for a Vessel of the Displacement Type.
PCT/NO2006/000073.
LARSSON, L. & RAVEN, H. C. 2010. Principles of Naval Architecture Series - Ship Resistance and
Flow. Society of Naval Architects and Marine Engineers (SNAME).
LAURSEN, W. 2001. Efficient New Bow Shapes Get Sexy. Baird Maritime [Online]. Available:
http://www.bairdmaritime.com/index.php?option=com_content&view=article&id=116
12&Itemid=208 [Accessed 20.05.2012].
LØVSTAD, M. 2008. Where iron ore is king - Port Complex of Tubarão [Online]. Available:
http://www.dnv.com/industry/maritime/publicationsanddownloads/publications/upd
ates/bulk/2008/2_2008/whereironoreiskingportcomplexoftubaro.asp [Accessed 22.03
2012].
MARUO, H. 1957. The Excess Resistance of a Ship in Rough Seas. International Shipbuilding
Progress, 4(35).
MARUO, H. 1960. The Drift of a Body Floating on Waves. Journal of Ship Research, 4 (3).
MATSUMOTO, K. 2002. "Ax-Bow" : A New Energy-saving Bow Shape at Sea. NKK Technical Review,
86, 46-47.
MATSUMOTO, K., NAITO, S., TAKAGI, K., HIROTA, K. & TAKAGISHI, K. Beak-bow to Reduce the
Wave Added Resistance at Sea. 7th International Symposium on practical design of ships
and mobile units., 1998 The Hague, The Netherlands. 527-533.
MINSAAS, K. & STEEN, S. 2008. Lecture Notes - Ship Resistance, Department of Marine Technology
- NTNU.
SAKAMOTO, T. & BABA, E. Minimization of resistance of slowly moving full hull forms in short
waves. Proceedings of the 16th Symposium on Naval Hydrodynamics, 1986 Berkely,
USA. 598-612.
SAUDI-ARAMCO. 2012. Ras Tanura Terminal [Online]. Available:
http://www.saudiaramco.com/en/home.html#our-
operations%257C%252Fen%252Fhome%252Four-operations%252Fports-and-
terminals%252Ffacilities%252Fras-tanura-terminal.baseajax.html [Accessed 22.03
2012].
SEA-JAPAN 2006. Universial Applies New Energy-aving Type Leadge Bow to 81,000DWT Bulker.
SEA-Japan, 318.
SEA-WEB. 2012. Fairplay Database [Online]. Available: www.sea-web.com [Accessed 18.04.2012
2012].
SIMMAN2008. 2008. Moeri Tanker KVLCC2 [Online]. Available:
http://www.simman2008.dk/KVLCC/KVLCC2/kvlcc2_geometry.html [Accessed 15.02
2012].
STEEN, S., DENG, G. B. & GUO, B. 2010. RANS Prediction and Experiment on Ship Resistance and
Flow Field of KVLCC2.
STEEN, S. & FALTINSEN, O. M. Added Resistance of a Ship Moving in Small Sea States. Practical
Design of Ships and Mobile units, 1998 Hague, The Netherlands.
STOPFORD, M. 2009. Maritime Economics, 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN,
Routledge.
SUNDE, C. A. 2011. The Importance of Added Resistance and Speed of Ships Due to Waves for Ships
Designed for Slow-Steaming. MSc, Norwegian University of Science and Technology.
TORSETHAUGEN, K. & HAVER, S. Simplified double peak spectral model for ocean waves. The
Fourteenth (2004) International OFFSHORE AND POLAR ENGINEERING CONFERENCE,
2004 Toulon, France. International Society of Offshore and Polar Engineers (ISOPE), 76-
84.
TUCK, E. O., SALVESEN, N. & FALTINSEN, O. M. 1970. Ship Motions and Sea Loads. Transactions of
the Society of Naval Architects and Marine Engineers, 78, 250-287.

64
TVETE, M. R. & BORGEN, H. 2012. A Ship's Fore Body Form. PCT/NO2010/000030.
UENO, M., NIMURA, T. & MIYAZAKI, H. On Steady Horizontal Force and Moment due to Short
Waves Acting on Ships in Maneuvring Motion. Proceedings of the 8th international
symposium on practical design of ships and other floating structure, 2001 Shanghai,
China. 671-677.
WILSON, P. A. 1985. A Review of the Methods of Calculation of Added Resistance for Ships in a
Seaway. Journal of Wind Engineering and Industrial Aerodynamics, 20, 187-199.

65
 Cummulative
Hs Tz
2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 sum
0.5 392089 1158575 2606822 3807976 3180799 1545938 475085 106601 20761 4258 1089 360 142 62 28 14 13300599
1 122828 459195 1392982 3157625 4166045 3161786 1473191 469398 115227 24499 5102 1194 351 130 57 27 27850236
1.5 52875 234320 819781 2343612 4075185 4150393 2568542 1056000 320749 79547 17654 3852 923 270 100 44 43574083
2 22271 113833 447147 1506650 3180309 4044938 3133058 1580128 571947 163411 40052 9080 2061 515 157 60 58389700
2.5 9234 52980 230730 882303 2143875 3242747 3033140 1823476 767665 248271 66949 16149 3719 879 233 76 70912126
3 3779 23772 113653 482151 1298977 2248268 2474601 1741409 839493 303134 89053 22838 5436 1276 318 92 80560376
3.5 1528 10353 53720 249708 726699 1394157 1767798 1437818 785734 313745 99512 26955 6634 1571 380 102 87436790
4 610 4405 24450 123990 383704 795799 1140978 1061619 653022 286502 97472 27707 7016 1676 400 102 92046242
4.5 241 1844 10746 59553 194858 429819 685485 722284 496967 238493 86696 25759 6682 1605 378 93 95007745
5 94 763 4572 27847 96626 225294 394551 465447 355535 185855 72053 22361 5939 1432 332 79 96866525
5.5 36 313 1888 12731 47278 116867 222979 290839 244235 138351 57130 18549 5061 1231 282 65 98024360
6 14 128 759 5705 22956 60663 125752 179236 163610 99750 43786 14905 4202 1038 237 53 98747154
6.5 5 52 299 2509 11087 31613 71234 109966 107941 70287 32688 11677 3417 866 200 44 99201039
7 2 21 115 1084 5327 16521 40518 67377 70525 48690 23887 8947 2725 712 168 37 99487695
Appendix A Scatter Diagram

I
7.5 1 8 44 460 2545 8639 23075 41208 45748 33292 17153 6721 2131 576 139 31 99669466
8 0 3 17 192 1208 4511 13121 25118 29483 22524 12141 4963 1636 458 114 26 99784981
8.5 0 1 6 79 569 2349 7436 15234 18872 15097 8488 3610 1235 358 91 21 99858427
9 0 0 2 32 266 1218 4194 9183 11988 10025 5867 2591 918 275 72 17 99905075
9.5 0 0 1 13 123 628 2353 5496 7553 6594 4010 1836 673 208 56 14 99934633
10 0 0 0 5 56 322 1311 3264 4715 4293 2710 1284 486 154 43 11 99953287
10.5 0 0 0 2 26 163 725 1922 2915 2765 1810 886 346 113 32 8 99965000
11 0 0 0 1 11 82 398 1121 1783 1761 1194 604 243 82 24 6 99972310
11.5 0 0 0 0 5 41 217 647 1079 1108 778 406 168 58 17 5 99976839
12 0 0 0 0 2 20 117 370 645 689 500 269 114 41 12 3 99979621
12.5 0 0 0 0 1 10 62 209 381 422 317 176 77 28 9 2 99981315
13 0 0 0 0 0 5 33 117 222 256 198 113 51 19 6 2 99982337
13.5 0 0 0 0 0 2 17 64 128 153 122 72 33 13 4 1 99982946
14 0 0 0 0 0 1 9 35 73 90 74 45 21 8 3 1 99983306
14.5 0 0 0 0 0 0 3 11 23 30 25 16 8 3 1 0 99983426

Sum 605607 2060566 5707734 12664228 19538537 21482794 17659983 11215597 5639019 2303892 788510 233925 62448 15657 3893 1036 99983426
This is posted as an example. Input files to ShipX can be found in the attached zip-file.

APPENDIX FIGURE 1 – WEIGHTED SCATTER DIAGRAM FOR THE WHOLE ROUTE FROM RAS TANURA TO LOOP
Appendix B Calculated Resistance Coefficients

APPENDIX FIGURE 2 - KVLCC2A

V [kn] Fn Rn Cf ΔCf - MARINTEK (1+k)*(Cf+ΔCf) C_DB Cw (wave cut) Cw (IP) Cr mod Ct mod
10.98428 0.1 1.55E+09 1.45E-03 1.06E-04 2.09E-03 8.52E-06 0.00E+00 0.00E+00 9.96E-04 3.10E-03
13.18113 0.12 1.86E+09 1.42E-03 1.37E-04 2.09E-03 8.62E-06 5.21E-06 9.09E-04 1.00E-03 3.10E-03
15.37799 0.14 2.17E+09 1.39E-03 1.62E-04 2.09E-03 8.70E-06 1.45E-05 8.14E-04 1.01E-03 3.11E-03
15.59767 0.142 2.20E+09 1.39E-03 1.64E-04 2.09E-03 8.70E-06 1.65E-05 8.08E-04 1.01E-03 3.11E-03
16.47642 0.15 2.32E+09 1.38E-03 1.73E-04 2.09E-03 8.73E-06 2.16E-05 7.97E-04 1.02E-03 3.11E-03
17.57484 0.16 2.48E+09 1.37E-03 1.83E-04 2.09E-03 8.77E-06 3.39E-05 7.88E-04 1.03E-03 3.13E-03
19.7717 0.18 2.79E+09 1.35E-03 2.01E-04 2.09E-03 8.83E-06 1.93E-04 9.54E-04 1.19E-03 3.28E-03

APPENDIX FIGURE 3 - KVLCC2B

V [kn] Fn Rn Cf ΔCf - MARINTEK (1+k)*(Cf+ΔCf) C_DB Cw (wave cut) Cw (PI) Cr mod Ct mod
10.9842781 0.1 1.55E+09 1.45E-03 1.06E-04 2.09E-03 8.31E-06 1.29E-06 5.73E-06 9.96E-04 3.10E-03
13.1811337 0.12 1.86E+09 1.42E-03 1.37E-04 2.09E-03 8.40E-06 1.56E-05 7.30E-05 1.01E-03 3.11E-03
15.3779893 0.14 2.17E+09 1.39E-03 1.62E-04 2.09E-03 8.48E-06 3.47E-05 1.23E-04 1.03E-03 3.13E-03
15.5976749 0.142 2.20E+09 1.39E-03 1.64E-04 2.09E-03 8.49E-06 4.03E-05 1.23E-04 1.03E-03 3.14E-03
16.4764171 0.15 2.32E+09 1.38E-03 1.73E-04 2.09E-03 8.52E-06 5.03E-05 1.40E-04 1.04E-03 3.14E-03
17.5748449 0.16 2.48E+09 1.37E-03 1.83E-04 2.09E-03 8.55E-06 9.37E-05 1.83E-04 1.09E-03 3.19E-03
19.7717005 0.18 2.79E+09 1.35E-03 2.01E-04 2.09E-03 8.61E-06 3.30E-04 4.50E-04 1.32E-03 3.42E-03

APPENDIX FIGURE 4 - KVLCC2C

V [kn] Fn Rn Cf ΔCf - MARINTEK (1+k)*(Cf+ΔCf) C_DB Cw (wave cut) Cw (PI) Cr mod Ct mod
11.118 0.1 1.61E+09 1.44E-03 1.07E-04 2.06E-03 8.54E-06 9.97E-07 1.76E-03 9.96E-04 3.07E-03
13.342 0.12 1.93E+09 1.41E-03 1.38E-04 2.06E-03 8.64E-06 5.72E-06 1.26E-03 1.00E-03 3.07E-03
15.566 0.14 2.25E+09 1.39E-03 1.63E-04 2.06E-03 8.72E-06 1.34E-05 9.58E-04 1.01E-03 3.08E-03
15.788 0.142 2.28E+09 1.39E-03 1.65E-04 2.06E-03 8.72E-06 1.65E-05 9.30E-04 1.01E-03 3.08E-03
16.678 0.15 2.41E+09 1.38E-03 1.74E-04 2.06E-03 8.75E-06 2.64E-05 8.49E-04 1.02E-03 3.09E-03
17.79 0.16 2.57E+09 1.37E-03 1.84E-04 2.06E-03 8.78E-06 4.73E-05 7.65E-04 1.04E-03 3.11E-03
20.013 0.18 2.89E+09 1.35E-03 2.02E-04 2.06E-03 8.84E-06 1.45E-04 6.73E-04 1.14E-03 3.21E-03

APPENDIX FIGURE 5 - KVLCC2D

V [kn] Fn Rn Cf ΔCf - MARINTEK (1+k)*(Cf+ΔCf) C_DB Cw (wave cut) Cw (PI) Cr mod Ct mod
10.9843 0.1 1.55E+09 1.45E-03 1.06E-04 2.09E-03 8.52E-06 2.41E-06 1.90E-04 9.96E-04 3.10E-03
13.1811 0.12 1.86E+09 1.42E-03 1.37E-04 2.09E-03 8.61E-06 9.55E-06 2.51E-04 1.00E-03 3.11E-03
15.378 0.14 2.17E+09 1.39E-03 1.62E-04 2.09E-03 8.69E-06 1.75E-05 2.86E-04 1.01E-03 3.11E-03
15.5977 0.142 2.20E+09 1.39E-03 1.64E-04 2.09E-03 8.70E-06 1.88E-05 2.95E-04 1.01E-03 3.11E-03
16.4764 0.15 2.32E+09 1.38E-03 1.73E-04 2.09E-03 8.73E-06 2.46E-05 3.24E-04 1.02E-03 3.12E-03
17.5748 0.16 2.48E+09 1.37E-03 1.83E-04 2.09E-03 8.76E-06 5.41E-05 3.74E-04 1.05E-03 3.15E-03
19.7717 0.18 2.79E+09 1.35E-03 2.01E-04 2.09E-03 8.82E-06 2.41E-04 6.77E-04 1.23E-03 3.33E-03

APPENDIX FIGURE 6 - KVLCC2E

V [kn] Fn Rn Cf ΔCf - MARINTEK (1+k)*(Cf+ΔCf) C_DB Cw (wave cut) Cw (PI) Cr mod Ct mod
11.251 0.1 1.66E+09 1.44E-03 1.09E-04 2.03E-03 8.42E-06 9.91E-07 1.57E-03 9.96E-04 3.03E-03
13.501 0.12 2.00E+09 1.41E-03 1.39E-04 2.03E-03 8.51E-06 5.15E-06 1.12E-03 1.00E-03 3.04E-03
15.751 0.14 2.33E+09 1.38E-03 1.64E-04 2.03E-03 8.59E-06 1.83E-05 8.56E-04 1.01E-03 3.05E-03
15.976 0.142 2.36E+09 1.38E-03 1.66E-04 2.03E-03 8.60E-06 2.39E-05 8.38E-04 1.02E-03 3.05E-03
16.877 0.15 2.50E+09 1.37E-03 1.75E-04 2.03E-03 8.63E-06 4.87E-05 7.87E-04 1.04E-03 3.08E-03
18.002 0.16 2.66E+09 1.36E-03 1.85E-04 2.03E-03 8.66E-06 1.00E-04 7.49E-04 1.09E-03 3.13E-03
20.252 0.18 3.00E+09 1.34E-03 2.02E-04 2.03E-03 8.72E-06 2.42E-04 7.26E-04 1.24E-03 3.27E-03

II
Appendix C RAO added resistance

APPENDIX FIGURE 7 - KVLCC2A: RAO ADDED RESISTANCE

8.00E+00
6.00E+00
4.00E+00 0 degrees
2.00E+00 45 degrees
Caw

0.00E+00 90 degrees
-2.00E+00 0 5 10 15 135 degrees
-4.00E+00 180 degrees
-6.00E+00
Lpp/λ

APPENDIX FIGURE 8 - KVLCC2B: RAO ADDED RESISTANCE

1.00E+01
8.00E+00
6.00E+00 0 degrees
4.00E+00 45 degrees
Caw

2.00E+00 90 degrees
0.00E+00
135 degrees
-2.00E+00 0 5 10 15
180 degrees
-4.00E+00
-6.00E+00
LPP/λ

APPENDIX FIGURE 9 - KVLCC2C: RAO ADDED RESISTANCE

1.00E+01
8.00E+00
6.00E+00 0 degrees
4.00E+00
45 degrees
Caw

2.00E+00
90 degrees
0.00E+00
135 degrees
-2.00E+00 0 5 10 15
180 degrees
-4.00E+00
-6.00E+00
LPP/λ

III
APPENDIX FIGURE 10 - KVLCC2D: RAO ADDED RESISTANCE

1.00E+01
8.00E+00
6.00E+00 0 degrees
4.00E+00
45 degrees
Caw

2.00E+00
90 degrees
0.00E+00
135 degrees
-2.00E+00 0 2 4 6 8 10 12 14
180 degrees
-4.00E+00
-6.00E+00
LPP/λ

APPENDIX FIGURE 11 - KVLCC2E: RAO ADDED RESISTANCE

1.00E+01
8.00E+00
6.00E+00
0 degrees
4.00E+00
45 degrees
Caw

2.00E+00
90 degrees
0.00E+00
135 degrees
-2.00E+00 0 5 10 15
180 degrees
-4.00E+00
-6.00E+00
LPP/λ

IV
Appendix D Attached Files in Zip-folder

Scatter diagram input files for ShipX:

AG-LOOP.sea – Arabian Gulf (Ras Tanura) to LOOP (USA)

AG-Chiba.sea - Arabian Gulf (Ras Tanura) to Chiba (Japan)

Mongstad-NY.sea – Mongstad (Norway) to New York (USA)

Tubarão-Quindao.sea – Tubarão (Brazil) to Quindao (China)

.igs-files, 3D models used in analyses:

KVLCC2_original.igs – Original KVLCC2 design

KVLCC2B.igs

KVLCC2C.igs

KVLCC2D.igs

KVLCC2E.igs