ISTANBUL TECHNICAL UNIVERSITY  GRADUATE SCHOOL OF SCIENCE

ENGINEERING AND TECHNOLOGY

STRUCTURAL ANALYSIS OF 9100 DWT

MULTI PURPOSE SHIP

M.Sc. THESIS

Barış DEDETAŞ

Department of Naval Architecture and Naval Machinery Engineering

Naval Architecture and Marine Engineering Programme

OCTOBER 2009
ISTANBUL TECHNICAL UNIVERSITY  GRADUATE SCHOOL OF SCIENCE
ENGINEERING AND TECHNOLOGY

STRUCTURAL ANALYSIS OF 9100 DWT

MULTI PURPOSE SHIP

M.Sc. THESIS

Barış DEDETAŞ
(508051016)

Department of Naval Architecture and Naval Machinery Engineering

Naval Architecture and Marine Engineering Programme

Thesis Advisor: Prof. Dr. Ahmet ERGİN

OCTOBER 2009
İSTANBUL TEKNİK ÜNİVERSİTESİ  FEN BİLİMLERİ ENSTİTÜSÜ

9100 DWT LİK ÇOK AMACLI GEMİNİN

YAPISAL ANALİZİ

YÜKSEK LİSANS TEZİ

Barış DEDETAŞ
(508051016)

Gemi İnşaatı ve Gemi Makinaları Mühendisliği Ana Bilim Dalı

Gemi İnşaatı ve Gemi Makinaları Mühendisliği Programı

Tez Danışmanı: Prof. Dr. Ahmet ERGİN

EKİM 2009
vi
FOREWORD

I would like to thank, Prof. Dr. Ahmet Ergin who ensure me a free working
circumstance, and also my family for everything.

OCTOBER 2009 Barış DEDETAŞ

(Junior Engineer)

vii
viii
TABLE OF CONTENTS

Page
FOREWORD ........................................................................................................vii
TABLE OF CONTENTS ....................................................................................... ix
ABBREVIATIONS ................................................................................................ xi
LIST OF TABLES .............................................................................................. xiii
LIST OF FIGURES .............................................................................................. xv
SUMMARY ......................................................................................................... xvii
ÖZET.................................................................................................................... xix
1. INTRODUCTION ............................................................................................... 1
1.1 Ship Manufacture and Scantling .................................................................... 1
1.2 Calculation of Strength via Empirical Formulation ........................................ 1
1.3 Calculation of Strength via Direct Strength Assessment ................................. 1
2. LITERATURE REVIEW ................................................................................... 3
2.1 Procedure, Method and Tools ........................................................................ 3
2.2 Modelling and Meshing ................................................................................. 3
2.3 Loading and Boundary Condition .................................................................. 5
3. FINITE ELEMENT METHOD ......................................................................... 7
3.1 Finite Eement Theory .................................................................................... 7
3.2 Coventional Pocket Programme ....................................................................10
4. STRUCTURAL ANALYSIS OF MULTI PURPOSE SHIP ........................... 11
4.1 Introduction ..................................................................................................11
4.1.1 Main dimensions ................................................................................. 12
4.2 Modelling Stage ...........................................................................................13
4.2.1 Solid modelling stage .......................................................................... 13
4.2.2 Mesh modelling stage ......................................................................... 16
4.2.3 Material properties .............................................................................. 19
4.3 Boundary and Loading Condition .................................................................19
4.3.1 Boundary conditions ........................................................................... 19
4.3.2 Loading conditions ............................................................................. 20
4.4 Results..........................................................................................................26
5. CONCLUSION AND RECOMMENDATIONS .............................................. 29
REFERENCES ..................................................................................................... 31
APPENDICES ....................................................................................................... 33
CURRICULUM VITAE ....................................................................................... 49

ix
x
ABBREVIATIONS

BV : Bureau VERITAS Classification Society, French

FEM : Finite Element Method
DWT : Deadweight of the ship
ps : Hydrostatic sea pressure.
g : Gravitational acceleration.
 : Density of sea water.
T : Drought of ship
L : Length of ship
B : Bred of ship
CB : Block coefficient.
CW : Water plane coefficient.
aB : Motion and acceleration parameter.
hW : Wave parameter.
C : Wave parameter.
H : Wave parameter.
M SWM , H : Design still water bending moment in hogging condition.
M SWM , S : Design still water bending moment in sagging condition.
M WV,H : Vertical wave bending moment in hogging condition.
M WV,S : Vertical wave bending moment in sagging condition.
n1 : Navigation coefficient.
CM : Wave torque coefficient.
CQ : Horizontal wave shear force coefficient.
QWV : Vertical shear force.
FM : Distribution factor.
FQ : Distribution factor.
FTM : Distribution factor.
FTQ : Distribution factor.
H : Heave acceleration.
R : Roll acceleration.
P : Pitch acceleration.
Y : Yaw acceleration.
TSW : Sway period.
TR : Roll period.
TP : Pitch period.

xi
AR : Roll amplitude.
AP : Pitch amplitude

xii
LIST OF TABLES

Page
Table 2.1 : Symetry condition at the model fore and aft ends. .............................. 6
Table 4.1 : Wave pressure for load cases C. ....................................................... 20
Table 4.2 : Wave pressure for load cases A and B. ............................................. 21
Table 4.3 : Internal pressures. ............................................................................ 21
Table 4.4 : Acceleration values. ......................................................................... 22
Table 4.5 : The moment values, applied directly to the model. ........................... 24
Table 4.6 : Load summary. ................................................................................ 25
Table 4.7 : Stess summary. ................................................................................ 27

xiii
xiv
LIST OF FIGURES

Page
Figure 2.1 : Application procedure of the analyses based on three dimensional
models. .............................................................................................. 4
Figure 2.2 : Model longitudinal extension ships more tahan 170m in length. ......... 4
Figure 2.3 : Model longitudional extension ships less than 170m in length. ........... 5
Figure 2.4 : Finite element modelling criteria. ....................................................... 5
Figure 3.1 : Element type. ..................................................................................... 9
Figure 4.1 : General arrangement. ........................................................................ 12
Figure 4.2 : Extension of the finite element model. .............................................. 13
Figure 4.3 : Solid model. ..................................................................................... 14
Figure 4.4 : Solid model (Bulkhead).................................................................... 14
Figure 4.5 : Solid model (Bottom). ...................................................................... 15
Figure 4.6 : Solid model (Side)............................................................................ 15
Figure 4.7 : Shell 63 element. .............................................................................. 16
Figure 4.8 : Mesh density of the finite element model. ........................................ 17
Figure 4.9 : Mesh density of the finite element model (Bulkhead). ...................... 17
Figure 4.10 : Mesh density of the finite element model (Side). .............................. 18
Figure 4.11 : Mesh density of the finite element model (Bottom). ......................... 18
Figure 4.12 : Boundary conditions. ....................................................................... 19
F
Figure 4.13 : FM , FTM , TQ values. ...................................................................... 24
Figure 4.14 : The summary of load cases............................................................... 25
Figure 4.15 : Displacement values of hatchcoaming. ............................................. 26
Figure A.1 : Shell thickness. ................................................................................. 33
Figure A.2 : Stress points. .................................................................................... 33
Figure B.1 : Sea pressure loading values of load case A (N/mm2). ........................ 34
Figure B.2 : Sea pressure loading values of load case B (N/mm2). ....................... 34
Figure B.3 : Sea pressure loading values of load case C (N/mm2). ....................... 35
Figure B.4 : Internal pressure loading values of load case A (N/mm2). ................. 35
Figure B.5 : Internal pressure loading values of load case B (N/mm2). ................. 36
Figure B.6 : Internal pressure loading values of load case C (N/mm2). ................. 36
Figure B.7 : Vertical bending moment distribution along the model at loading
condition A. ..................................................................................... 37
Figure B.8 : Vertical bending moment distribution along the model at loading
condition B....................................................................................... 37
Figure C.1 : Von misses stress result at load case A (MPa). .................................. 38
Figure C.2 : Y direction displacement at load case A (mm). ................................. 38
Figure C.3 : YZ direction share stress result at side girders at load case A
(MPa). .............................................................................................. 39
Figure C.4 : XY direction share stress result at side girders at load case A
(MPa). .............................................................................................. 39
Figure C.5 : Von misses stress result at side girders at load case A (MPa). ........... 40
Figure C.6 : Von misses stress result at double bottom at load case A (MPa). ...... 40

xv
Figure C.7 : YZ direction share stress result at double bottom at load case A
(MPa). ............................................................................................. 41
Figure C.8 : XZ direction share stress result at side girders at load case A
(MPa). ............................................................................................. 41
Figure C.9 : Von misses stress result at load case B (MPa). ................................. 42
Figure C.10 : Y direction displacement at load case B (mm). ................................. 42
Figure C.11 : YZ direction share stress result at side girders at load case B
(MPa). ............................................................................................. 43
Figure C.12 : XY direction share stress result at side girders at load case B
(MPa). ............................................................................................. 43
Figure C.13 : Von misses stress result at side girders at load case B (MPa)............ 44
Figure C.14 : Von misses stress result at load case A (MPa). ................................. 44
Figure C.15 : YZ direction share stress result at double bottom at load case B
(MPa). ............................................................................................. 45
Figure C.16 : Von misses stress result at double bottom at load case B (MPa). ...... 45
Figure C.17 : Von misses stress result at load case C (MPa). ................................. 46
Figure C.18 : Y direction displacement at load case C (mm). ................................. 46
Figure C.19 : Von misses stress result at side girders at load case C (MPa)............ 47
Figure C.20 : XY direction share stress result at side girders at load case C
(MPa). ............................................................................................. 47
Figure C.21 : Von misses stress result at double bottom at load case C (MPa). ...... 48
Figure C.22 : XY direction share stress result at side girders at load case C
(MPa). ............................................................................................. 48

xvi
 STRUCTURAL ANALYSIS OF 9100 DWT MULTI PURPOSE SHIP

SUMMARY

Nowadays, with the help of rapid improvements in technology, more complex and
much bigger scale problems are become solvable with respect to past. As a good
example, optimum design of ship structures, which is a very though problem, can be
solved with applying numerical method especially finite element method with the
help of computer technology.
The ship has been designed, depending on class society’s rules. These rules depend
on the class societies experiences. Finite element method as a direct strength
assessment is becoming a scantling method in the rule book, but it will take time to
use these methods effectively, and efficiently.
In today’s rules FEM analysis procedure of the ships, which are 170m in length, is
described and these are almost enough for every kind of analysis, but same thing is
not valid for ships, which are less than 170m in length.
Main aim of this work is to offer a FEM analysis procedure for small, which are less
than 170m in length ships by means of strength, and analyzing the 9100 DWT multi
purpose ship with conventional finite element package program.
The ship, which we analyzed, has large deck opening with two cargo hold and its
length is less than 170m. Therefore, we can not use the direct strength assessment
method based on three cargo hold model. Because of the ship has large deck
openings, it is expected that the maximum deflection of the hatchway coamings
ocuures at the middle of the hatch. In one cargo hold model, the ships are modelled
from the middle of the one cargo hold to the middle of the adjacent cargo hold,
including the bulkhead. The boundary conditions are applied to the end of the model
as well. Applying boundary condition to the place where the maximum displacement
occurs is not right way for this kind of analysis. For this reason a new procedure was
developed.
In this work, I tried to carry out structural analyses of a 9100 DWT multi purpose
ship with conventional finite element package program. When analyzing this ship, a
direct calculation method, which has originally developed by the classification
societies, has been used with some modification in modelling a boundary condition,
and result has been presented.

xvii
xviii
 9100 DWT LİK ÇOK AMACLI GEMİNİN YAPISAL ANALİZİ

ÖZET

Günümüzde bilgisayar ve yazılım sektöründeki hızlı ilerlemeler sayesinde büyük

boyutlu ve karmaşık yapılı problemlerin çözümleri de kolaylaşmaktadır. Buna güzel
bir örnek ta gemi imalatının, boyutlandırma ve dizayn aşamalarında, sonlu elemanlar
yöntemi ile yapılan analizlerdir.
Sonlu elemanlar yöntemi yardımıyla yapılan analizler sanayinin diğer alanlarında
daha yaygın olarak kullanılırken, gemi inşa sanayisinde kullanımı pek yaygın
değildir. Bunun en önemli nedeni gemi inşam sanayinin ülkemizde ve dünyada
geleneksel “tecrübeye dayalı” yöntemler üzerine kurulu olması ve de bu sanayinin
ana unsuru olan Sınıflandırma “Klas” kuruluşlarının ataletidir.
Günümüzde, halihazırda inşa edilen gemiler bir klas kuruluşunun kurallarına bağlı
kalarak inşam edilmektedir ve bu kurallar sınıflandırma kuruluşlarının “çoğu yüz yılı
aşan” tecrübe ve birikimlerinin bir sonucudur. Sonlu elemanlar yönteminin kullanımı
direk hesap yöntemi adı altında, sınıflandırma kuruluşlarının kuralları arasında yerini
alsa da, bu kuralların işlerlik kazanması ve prosedürlerin gelişmesi zaman alacaktır.
Halihazırda 170m ve üzeri boylarda sonlu elemanlar yöntemiyle analiz prosedürü
tanımlanmış iken, boyu 170m’nin altında olan gemiler için olan analiz prosedürü
yetersiz kalmaktadır.
Bu çalışmada temel amacımız, gemi boyutlandırmasında sonlu elemanlar
kullanımına bir prosedür önermek, ve bu önerilen prosedür çerçevesinde 9100
DWT’lik çok amaçlı bir kuru yük gemisinin yapısal analizlerini gerçekleştirmektir.
İncelemiş olduğumuz gemi, geniş ambar ağzı açıklıklı ve iki ambarlıdır. Hem iki
ambarının olması hem de 170m den küçük olması nedeniyle sınıflandırma
kuruluşlarının “burada BV sınıflandırma kuruluşunun kuralları baz alınmıştır” 170m
üzeri gemiler için önerdiği gemi ortasındaki üç ambarın modellemesine dayalı direk
hesap prosedürü kullanılamamaktadır. Tek ambarın modellemesine dayalı hesap
prosedürü (gemi ortasına denk gelen ambar perdesi ile bu perdenin komşuluğundaki
ambarların ortalarına kadar olan kısmın modellenmesi ile yapılan analiz) de bu tipte
bir gemi için yetersiz kalmaktadır.
Tek ambar analizinin yetersiz kalmasının en önemli nedeni; geminin ambar ağzı
açıklıklarının fazla oluşu, bu nedenle de, yükleme durumuna bağlı olarak, ambar ağzı
yer değişimlerinin en yüksek değerlerinin ambar ortalarında olmasının
beklenmesidir. Sınır koşullarının maksimum yer değişmelerin olduğu bölgelere
uygulanması sağlıklı bir analiz yapılabilmesi açısından sakıncalar içermektedir.
Bu analizin neticesinde ambar ağzı yer değişimlerinin tespit edilerek, bu bulunan
değerler doğrultusunda bu yer değişmeleri tolere edecek şekilde ambar kapağı
dizaynı önerilebilir. Bu durumda ambar ağzı deplasmanlarının doğru tespiti bir kat
daha önem kazanmaktadır

xix
Özet olarak, bir prosedür önerilmiş ve bu önerilen prosedür doğrultusunda analizler
yapılmıştır. Yapılan analizlerin neticesinde ambar ağızlarında oluşan yer
değiştirmeler bulunmuştur. Ayrıca gemi üzerindeki belirli noktalarda bulunan
gerilme değerleri de sonuç kısmında belirtilmiştir.
Ayrıca tez içerisinde sonlu elemanlar teorisine matematiksel anlamda değinilmiş,
analizde kullanılan paket program hakkında da bilgi verilmiştir, analizde kullanılan
eleman formülasyonu matematiksel ve geometrik olarak ifade edilmiştir.
Yapısal analiz aşamasında geminin boyutları ve geometrik özellikleri verilmiş ve bu
geminin ANSYS programı içerisinde nasıl modellendiği ayrıntılı olarak anlatılmıştır.
Gemi 9100 DWT kapasiteli, tam boyu 131.07m, kalıp genişliği 17.20m, su çekimi
7.40m, derinliği 9.80m ve servis hızı 13 knot olan çok amaçlı bir kuru yük gemisidir.
Analizi yapılan gem iki ambarlı çok amaçlı bir kuru yük gemisidir. Geminin
konteyner taşıma durumu da olduğundan ambar ağzı açıklıklarının geniş olması ve
iki ambarlı olması analizi son derece hassas ve de sıkıntılı bir hale sokmaktadır.
BV kurallarında önerilen, üç ambar modelleyerek yapılan analiz prosedürü, analizini
yapmak istediğimiz geminin üç ambarı bulunmamasından dolayı, tek ambar analizi
de, geminin ambar ağzı açıklıklarının fazla olması ve de sınır koşullarının ambar
ortalarına (deformasyonun en fazla olması beklenen bölge) uygulanmasının
doğuracağı sakıncalar nedeniyle uygulanmamaktadır. Bu durumda değişik bir
prosedür olarak geminin mevcut iki ambarı modellenmek ve de sınır koşulları da
ayrıca hesaplanıp modelin uçlarına uygulanmak suretiyle analiz yapılmıştır.
Geminin baş çatışma perdesinden makine dairesi perdesine kadar olan kısmı
modellenmiştir. Modelleme daha öncede belirtildiği gibi ANSYS paket programı
içerisinde yapılmış hollanda profilleri dahil bütün elemanlar kabul (Shell) eleman
olarak modellenmiştir. Elemanın matematiksel formulasyonu ayrıntılı olarak
verilmiştir. Kullanılan eleman ANSYS programı içerisindeki Shell 181 dir ve daha
ayrıntılı bilgi için ANSYS kullanım kılavuzundan faydalanılabilir.
Modele sac kalınlıkları atanırken yine BV kuralları gereği çeşitli bölgelerdeki
korozyon artımları düşülmek suretiyle net kalınlıklar kullanılmıştır.
Model tamamlandıktan sonra sonlu elemanlar ağı oluşturma, kısaca mesh işlemine
geçilmiştir. Modelin iki ambar arasında kalan perde ve bu perdeden baş ve kıç
doğrultusunda ambar ortalarına kadar olan kısım daha sık ve düzgün meshlenmiş (
BV kurallarına göre fine mesh diye tabir edilen,dörtgen elemanın bir kenarı postalar
arası mesafenin üçte biri olacak şekilde), ambar ortalarından baş kıç doğrultuda baş
çatışma perdesi ve makine dairesi perdesine kadar olan kısım daha seyrek(corse
mesh) meshlenmiştir.
Mesh işleminden sonra sonra sınır koşullarının atanmasına geçilmiştir. Geminin orta
kesit tarafsız ekseninin merkez hattı ile kesişim noktasında bir düğüm noktası
oluşturulmuş ve modelin baş ve kıç kesitindeki düğüm noktaları (nodlar) bütün
serbestlik derecelerinde bu belirlenen tarafsız eksendeki düğüm noktası birlikte
hareket edecek şekilde sınırlanmıştır (coupling). Daha sonra gemiyi bir kiriş olarak
kabul eden analitik çözümden elde edilen deplasman, kuvvet, ve moment değerleri
belirlenmiş bu değerler klas kuruluşunun ampirik formüllere bulunan değerleri ile
mukayese edilmek ve birleştirmek suretiyle modelin iki ucundan sabitlenen düğüm
noktalarına sınır koşulu olarak atanmıştır.

xx
Yükleme koşullarında ise bu sınır koşullarına ek olarak ,zira bu sınır koşulları da
kendi başlarına yükleme içermektedir, yine BV kuralları baz ambar içerisine ve
denize batan kısmına hidrostatik basınç uygulanmıştır. Uygulanan bu basınç
değerleri aslında dinamik etkileri de içinde barındırmaktadır, zira gemi hareketinden
kaynaklanan ivme değerleri de bu basınçlar uygulanırken dikkate alınmıştır.
Yapılan analizler neticesinde elde edilen gerilme ve yer değiştirme değerleri tablo ve
görsel olarak sunulmuştur. Sonuçlar itibariyle maksimum yer değiştirme değerlerinin
140 mm ile büyük ambarın ortasında oluştuğu görülmüştür.
Ayrıca geminin çeşitli bölgelerinde ve elemanlarında oluşan Von Misses bileşik
gerilmeleri ve kayma gerilmeleri değer olarak sunulmuş görsel olarak ifade
edilmiştir. Buradan da gerilme değerlerinin klas kuruluğunun güvenlik gerilmeleri
sınırlarını aşmadığı görülmüştür.

xxi
1. INTRODUCTION

1.1 Ship Manufacture and Scantling

The structural analyse, in other word scantling is the most important phenomena and
stages of the ship design. Because of the designer always try to make a ship which
has maximum cargo capacity, and minimum light weight. The calculation of wave
load and load combinations is the first step in marine structural design. For structural
design and analyses, a structural engineers needs to have basic concept of waves,
motion and design loads [1].

Once the functional requirements and loads are determined an initial scantling may
be sized based on formulae and charts in classification rules and design codes. Basic
scantling of the structural components is initially determined based on stress analyses
of beams plates and shells under hydrostatic pressure, bending and concentrated
loads [1].

1.2 Calculation of Strength via Empirical Formulation

The classification societies have a great experience and their own database related to
the every kind and size ships. Using their database and experience they are
developing some empirical formulation for scantling of ships. Although these
formulations are not based on direct calculation, these also reflect a good approach to
the scantling.

1.3 Calculation of Strength via Direct Strength Assessment

The traditional design rule formulae involve a number of simplification, assumptions

and can only be used in certain limits. Moreover, scantlings based on rules are not
necessarily the most cost efficient designs. Hence, the application of rational stress
analysis using FEM has gained increasing attention in the shipbuilding industry.

1
2
2. LITERATURE REVIEW

Classification rules have been the main stage of ship design. These rules are
primarily semi empirical in nature and have been calibrated to ensure successful
operational experience. But with the development of the computation technology, the
calculation methods of class societies have been developed accordingly. Nowadays
almost all of the class societies have their own procedure for direct calculation
method. Although, they are almost same, we explain in this chapter that the direct
calculation procedure of BV.

2.1 Procedure, Method and Tools

This procedure deals with the part of the structural analysis which aims at,
calculating the stress in the primary supporting members in the midship area, and
when necessary, in other areas which are to be used in yielding and buckling checks.

All primary supporting members in the midship region are to be included in three
dimensional model, with purpose of calculating their stress level and verifying their
scantlings. The application procedure of the analyses based on three dimensional
models can be seen in the Figure 2.1

2.2 Modelling and Meshing

The analysis of primary supporting members is to be carried out on fine mesh model.
The longitudinal extension of the structural model is to be such that, the hull girder
stress in the area to be analysed are properly taken into account in the structural
analysis, the result in the areas to be analysed are not influenced by the unavoidable
inaccuracy in the modelling of the boundary conditions.

In general for multi hole ships more than 170 m in length, the condition mentioned
above are considered as being satisfied when the model extended over at least three
cargo hold which can be seen in Figure 2.2

There are some criteria of finite element modelling according to the BV rules.

3
• The analysis of primary supporting members based on fine mesh models is to
be carried out by applying one of the following procedures, depending on the
computer resources:

• an analysis of the whole three dimensional model based on a fine

mesh

• an analysis of the whole three dimensional model based on a coarse

mesh, from which the nodal displacements or forces are obtained to be
used as boundary conditions for analyses based on fine mesh models
of primary supporting members [14].

Figure 2.1 : Application procedure of the analyses based on three dimensional

models.

Figure 2.2 : Model longitudinal extension ships more then 170m in length.

4
For ships less than 170 m in length, the model may be limited to one cargo hold (one
half cargo hold length on either side of the transverse bulkhead.) which can be seen
in Figure 2.3.

Figure 2.3 : Model longitudinal extension ships less than 170m in length.

Figure 2.4 : Finite element modelling criteria.

2.3 Loading and Boundary Condition

The hydrostatic water pressure is to be applied to the outside hull with increasing the
gravity as acceleration calculated from related class rules. For cargo hold, cargo load

5
weight are to be applied increasing the gravity same as sea pressure. And Global
bending moment and sheering forces are to be applied at the end of the model in
order to ensure maximum bending moment at the middle of the ship [14].

Structural model extended over at least three cargo tank/hold lengths; the whole three
dimensional model assumed to be fixed at one end, while shearing force, and
bending moments are applied at the other end to ensure the equilibrium. At the free
end section, rigid constraint conditions are to be applied to all nodes located on
longitudinal members, in such a way that the transverse section remains plane after
deformation [14].

Structural models extended over one cargo tank/hold length Symmetry conditions are
to be applied at the fore and aft ends of the model, as specified in Table 2.1.

Table 2.1 : Symmetry condition at the model fore and aft ends.
DISPLACEMENTS ROTATION
in directions around axes
X Y Z X Y Z
fixed free free free fixed fixed

6
3. FINITE ELEMENT METHOD

3.1 Finite Element Theory

The finite element method is basically a numerical method for solving differential
equation. The finite element method generates discrete models for continuous
system. The basic concept of the finite element method is the same as in matrix
frame analyse, that the structure can be represented as an assembly of individual
structural elements interconnected at discrete number of nodes. Continuous structure
such as panel of plates,

The First step creating finite element model for solid objects or structure is to divide
into finite number of discrete elements. There are usually selected from a library of
element types available within a given program. The element which is selected for
this work was described in chapter 4.

Each element is characterized by its own topology (an order of sequences of point or
nodes.), and by a number of material or structural property such as density, elasticity
module and young module.

Elements which carry loads in bending such as beams and plates are termed as
structural elements. They are more complex, since their formulation incorporate
aspects of beam and plate theory such as “plane section remain plane “, and involves
curvature and moments rather than simple stress and strain.

The form of the finite element equation can be described simply as; the displacement
at any point in structure due to the combined application of all external loads is
equivalent to the sum of the displacement at the same point due to the application of
the load separately, moreover the displacement at any point due to the application of
each load varies linearly with the magnitude of load.

These statements are given mathematical expression by defining quantity as the

contribution to the displacement which results from the application of a nodal force
in the absence of other loads. So it may be written as:

7
 1  11  12  13 .......  1n
 2   21   22   23 .......   2 n
. (3.1)
.
 n   n1   n 2   n3 .......   nn

These statements also implies the existence of a constant of proportionality which

determines the contribution made by the load f j to the displacement  ij .This

constant is termed the “flexibility influence coefficient” and defined as:

 ij  cij f j (3.2)

Leaving aside, the problem of actually calculating these quantities, we can substitute
equation 3.2 into 3.1 to produce a system of linear equations which relate the nodal
displacement to the nodal forces. These are;

1  c11 f1  c12 f 2  .......  c1n f n

 2  c21 f1  c22 f 2  .......  c2n f n
. (3.3)
.
 n  cn1 f1  cn 2 f 2  .......  cnn f n

They may be expressed more concisely in matrix form as;

d  Cf (3.4)

Where;

1   c11c12 .............c1n   f1 

   c c .............c  f 
 2  21 22 2n   2
.  .  . 
d    ,C   , f    (3.5)
.  .  . 
.  .  . 
     
 n   cn1.................cnn   f n 

The matrix C has n rows and n columns and is termed the flexibility matrix. Matrix f
and d are termed the nodal force vector and the nodal displacement vector,
respectively.

8
To understand the structure of the equivalent finite element equations, it is helpful to
consider the equations which would result if equation 3.4 were inverted, both sides of
the equation 3.4 can be multiplied by inverse of C to give;

C 1d  f (3.6)

The matrix C 1 is the “stiffness matrix” of the system. It is usually denoted by the
symbol K, and above equation is written.

Kd  f (3.7)

In finite element formulation, the coefficients of the stiffness matrix K are obtained
directly from the structure without reference flexibility coefficient.

The solution of equation 3.7 gives the nodal displacement of structure. After these
have been found, the strain within each element can be determined as well. The finite
element model represents, in this sense, a displacement method of analyse, because
the displacement of the structure are the primary unknown quantities and are
calculated first.

Figure 3.1 : Element type.

9
This is very basic explanation of finite element method, depending on the element
selection it may be more complex form. Although it is main aim to give a new kind
of strength calculation method by conventional pocket program for this paper,
theorical explanation of method was kept simple.

3.2 Conventional Pocket Programme

When we create model in conventional package programme, like Ansys, first of all
we create a solid model of structure that will be analysed, and than the solid model is
converted to the finite element model via mesh generation. This transformation
ensures us to define our model by nodes. These nodes are converted to the stiffness
matrix by using a special algorithm at background, depending on used coordinate
axis as well. Applied loads are also converted to the load vector. After the
generation of stiffness matrix and load vector, the equation is established, and
transferred to the solver. There is a linear algebraic equation, and can be solved by
arbitrary method.

All of the conventional package programme use almost same style except modelling
stages. We used Ansys in order to carry out our structural analyse. The analyses
sequence was given in chapter 4.

10
4. STRUCTURAL ANALYSIS OF MULTI PURPOSE SHIP

4.1 Introduction

The aim of this analysis is to investigate the structural endurance of main cargo tanks
construction of 9100 DWT Multi Purpose Ship. For this reason two cargo hold and
the bulkhead between two cargo hold have been modelled. Basically BV rules have
been taken into account for calculation. The analysis procedure has been explained in
chapter 2. Although, basically we follow these procedures, we will make a special
procedure.

The ship which we analysed has two cargo hold and 130m in length. Therefore we
can not apply three cargo analysis procedure. We carried out one cargo hold analysis
procedure by modelling two cargo hold without end bulkhead, including middle
bulkhead and finely meshing middle part ( from the middle of the first cargo hold to
the middle of the second cargo hold ), and coarsely meshing the other part of the
mode. This kind of modelled has been chosen in order to apply boundary condition
more correct, and handle more correct solution at the middle of the cargo holds.
Because if we apply boundary condition at the middle of the cargo holds, we can not
handle good solutions at these areas.

As a result we applied one cargo hold procedures by taking the load values from
class rules, which is presented in section 4.3 detaily, with modelling two cargo hold.
Although it is out of class procedure, we think it is better to use these kind of
analysis method to ensure more accurate solution.

First of all we calculated pressure values for our model from class rules, than these
pressure values were applied to the model, and the maximum moment value which
occurred on the model were calculated. Than we calculated moment values from the
class rules and found the difference between the moments occurred in the model and
calculated from class rules, and by adding these difference moment values to the
model, we shifted the moment of the model correct value. Same procedures were
applied for shearing forces.

11
Model was fixed at both ends as applying rigid region with base point which is
placed at the natural axis, and these difference moment values were applied to these
points. For tensional moment, which is in the load case C, there is a point which
placed at shear centre, and the tensional moment was applied from this point.

Then, the analyses were carried out, and results were presented.

4.1.1 Main dimensions

Length Overall 131.07m

Length Between P.P 123.20m

Breadth Moulded 17.20m

Depth Moulded 9.80m

Scanting Draught 7.40m

Deadweight 9100ton

Speed 13knot

Figure 4.1 : General arrangement.

12
4.2 Modelling Stage

The ship has two cargo tanks; although it is aimed that one cargo tank analyzed, all
of the cargo tanks has been modelled, including middle bulkhead and excluding aft
and fore bulkhead, because of applying more accurate boundary condition. The
extension of the model can be seen in Figure 4.2 below.

Figure 4.2 : Extension of the finite element model.

4.2.1 Solid modelling stage

The solid model is prepared in Ansys software and can be seen in Figures 4.3 to 4.5
below. In these Figures different colours indicate different thickness values and the
detailed outputs of thickness values can be seen in Appendix I. In the definition stage
of the thickness values, the corrosion additions are also taken into account according
to BV Rules, Part B, Chapter 4, Section 2, [3] Table 4.2. Thickness values can be
seen in Appendix I without corrosion additions. The ship has two cargo hold as can
be seen figure Figure 4.1. Both cargo holds have been modelled with all primary
structural members, such as longitudinal frames, girders, and webs with manhole and
face plates, bulkhead excluding fore most, and aft bulkheads. End bulkhead have
been excluded because, we have applied the boundary condition from these
locations.

13
 Figure 4.3 : Solid model.

Figure 4.4 : Solid model (Bulkhead).

14
Figure 4.5 : Solid model (Bottom).

Figure 4.6 : Solid model (Side).

15
4.2.2 Mesh modelling stage

Ansys solid model is meshed by using the Ansys specific elements Shell63. This
element is used for all plating parts, and profiles. The properties of this element are
given below Figure 4.7.

Figure 4.7 : Shell 63 element.

SHELL63 has both bending and membrane capabilities. Both in-plane and normal
loads are permitted. The element has six degrees of freedom at each node:
translations in the nodal x, y, and z directions and rotations about the nodal x, y, and
z axes.

Stress stiffening and large deflection capabilities are included. A consistent tangent
stiffness matrix option is available for use in large deflection (finite rotation)
analyses. This element can be seen in Figure 4.7.

After the meshing operations are completed, it is seen that the finite element model
has 195449 nodes and 211187 elements in total. The finite element model can be
seen in Figures 8 to 11.

Because of the fact that, we focused on the middle of the ship (from the middle of the
first tank to the middle of the second tank) the finite element model is considered as
finely meshed according to the BV rules Pt B, Ch 7, App. 1 3.4.3. at this zone (from
the middle of the first tank to the middle of the second tank). The other part of the
model is considered as course mesh.

16
 Figure 4.8 : Mesh density of the finite element model.

Figure 4.9 : Mesh density of the finite element model (Bulkhead).

17
 Figure 4.10 : Mesh density of the finite element model (Side).

Figure 4.11 : Mesh density of the finite element model (Bottom).

18
4.2.3 Material properties

The material for the steel used in the ship is St 42, AH 36 grade shipbuilding steel
and their properties are given below

• Elasticity modulus = 210000 N/mm2

• Poisson ratio = 0.30
• Density = 7850 kg/m3

4.3 Boundary and Loading Condition

4.3.1 Boundary conditions

The nodes at the aft and fore end cross sections are coupled to section’s centre nodes
near to the natural axis and simple supported boundary conditions (ux, uy, uz, Rx, Rz
are fixed for aft end and uy and uz are fixed for fore end) are applied for loading
condition A,B, and C. These centre nodes are also to be used to apply the global
bending and wave induced moment values. In load case C Rz was not fixed in order
to applying horizontal bending moment correctly, and for applying torsional moment
additional key point has been created. Boundary conditions can be seen in Figure12
below.

Figure 4.12 : Boundary conditions.

19
4.3.2 Loading conditions

It is aimed to get the worst loading condition for ship structure. Therefore load case
A B, and C according to the BV rules were applied to the model, lightship weight
was also included in these load case as double bottom load. The quasi static analysis
has been carried out. It means; static analysis was carried out by taking in the
account the dynamic effects by including the acceleration. The loads applied to the
model can be classified in four groups, such as; sea pressure, internal pressure,
moments, and shear forces.

Sea pressure is composed by two parts; still water pressure (hydrostatic pressure
applied by se to ship hull), and wave pressure.

Sea water pressure at any point of the hull is obtained from the formula as follows;

ps   g (T  z ) (4.1)

z is the point which you want the learn the pressure value, and if it is chosen grater
than T, p s is equal to 0.

Wave pressure depending on the load case can be calculated from the formula given
in the Tables 4.1 and 4.2.

Table 4.1 : Wave pressure for load cases C.

20
 Table 4.2 : Wave pressure for load cases A and B.

Internal pressure is composed by two parts; cargo hold load with increasing the
gravitational acceleration as corresponding calculated acceleration according to the
load cases, and light weight load.

The still water and inertial pressure are obtained in kN/ m 2 as specified in table 4.3.

Table 4.3 : Internal pressures.

21
With using formulas for aB and hW given below and table 4.4 the acceleration values

can be calculated.

hW
aB  n1 (0.76 F  1.875 ) (4.2)
L
L  250
hW  11, 44  for; L<350 (4.3)
110

Table 4.4 : Acceleration values.

Moments and shearing forces are calculated directly from the rules as below
explained.

M SWM , H  175n1CL2 B(CB  0, 7)103  M WV , H (4.4)

M SWM , S  175n1CL2 B(CB  0, 7)103  M WV , S (4.5)

C is the wave parameter, and calculated as follows;

300-L 1,5
C=10,75-( ) for 90m|<L<300m (4.6)
100

n1 is the navigation coefficient, and it is taken 1 for unrestricted navigation.

22
The vertical wave bending moments at any hull transverse section are obtained from
the following formula in k Nm.

Hogging conditions:

M WV,H = 190 FM n1CL2 BCB10-3 (4.7)

Sagging conditions:

M WV,S = 110 FM n1CL2 B(C B +0,7)10-3 (4.8)

The horizontal wave bending moment at any hull transverse section are obtained
from the following formula in k Nm.

M WH  0, 42 FM n1 HL2TCB (4.9)

H is the wave parameter, and calculated as follows;

250  0, 7 L 3
H  8.13  ( ) (4.10)
125

The values of the wave torque at any transverse section are calculated, with respect
to section centre of torsion, with following formula.

HL
M WT  n1 ( FTM CM  FTQCQ d ) (4.11)
4

CM is wave torque coefficient, and calculated as follows;

CM  0, 45 B 2 CW 2 (4.12)

CQ
is horizontal wave shear coefficient, and calculated as follows;

CQ  5TC B (4.13)

d is the vertical distance, in m, from the centre of torsion to a point located at 0,6T
above the base line.

The vertical shear forces at any hull transverse section are obtained from the
following formula in k N.

QWV  30 FQ n1CLB (CB  0, 7)102 (4.14)

FM , FTM , FTQ values can be taken from the Figure 4.13 below.

23
The moment values, applied directly to the model can be calculated using table 4.5.

FM , FTM , FTQ values.

Figure 4.13 :

Table 4.5 : The moment values, applied directly to the model.

The summary of these load cases can be seen from the Figure 4.14 below.

The loads, which are applied to the model directly, as result of presented calculation,
and explained shifting operation, are summarized in the Table 4.6. All the formulas
given above have been taken from BV rules Section B Chapter 5, and detailed
explanation of this formulation can be found in this Rule Book.

24
 Figure 4.14 : The summary of load cases.

Table 4.6 : Load summary.

Loading Cases

Sea Load A B C

Double Bottom Pressure (kN/m2) 101 140 113

Msw -180000 -180000 -180000

Mwv 342000 -368000 -368000

Maximal Vertical Bending Moments ( kNm) 55125 -433000 -281200

Maximal Horizontal Bending Moments

(kNm) 0 0 -253000

Torsional Moment (kNm) 50000

25
4.4 Results

In this work I tried to carry out structural analyses of a 9100 DWT multi purpose
ship. The subjected vessel has large deck openings, because of that we focused on
the deflections, which occur on the way of hatchway coamings. These deflections
value are important also for the manufacturing of the hatch cover. These deflection
values must be taken in to account, when manufacturing of the hatch cover.

The displacement values of the hatchcoaming of cargo hold No:2 which is in

between frame number 32 to frame number 105 are presented in Table 4.7 below.
The displacement values are absolute value in this table. In load case A and B
displacements are through inside the cargo hold, and in load case C displacements
are through outside the cargo hold.

As a result of this analysis, it has been seen that the plate which are AH and A grade
do not exceed their strength limits (230 Mpa for A grades, 360 Mpa for AH
grades).The stress values which occur on the plate because of the loading condition
can be seen on related pictures in Appendix II, and the stress values, corresponding
the point number declared in the Figure 2 of Appendix I, are summarized in Table
4.7. The shearing stress values were given as absolute values.

Displacement Values of Hatchcoaming

140
120
Displacements (mm)

100
Load Case A Deflections
80
Load Case B Deflections
60
Load Case C Deflections
40
20
0
65 75 85 95 105
Frame Number

Figure 4.15 : Displacement values of hatchcoaming.

26
 Table 4.7 : Stress summary.

Point Load Cases

A B C

Displacement 58,13 mm 54,91mm 122.41mm

1 Von Misses 162.95 N/mm2 140.05 N/mm2 102.05 N/mm2

Shear (XY) 22.52 N/mm2 20.35 N/mm2 17.8 n/N/mm2

Von Misses 282.45 N/mm2 201.76 N/mm2 204.65 N/mm2

2 Shear (XZ) 26.53 N/mm2 19.62 N/mm2 18.9 N/mm2

Von Misses 197.86 N/mm2 150.02 N/mm2 153.09 N/mm2

3 Shear (YZ) 79.882 N/mm2 73.296 N/mm2 125.22 N/mm2

Von Misses 175.15 N/mm2 200.54 N/mm2 221.20 N/mm2

4 Shear (XY) 68.51 N/mm2 71.44 N/mm2 70.75 N/mm2

Von Misses 113.87 N/mm2 152.27 N/mm2 122.42 N/mm2

5 Shear (XZ) 80.01 N/mm2 92.02 N/mm2 116.28 N/mm2

27
28
5. CONCLUSION AND RECOMMENDATIONS

In this work, it is aimed to create a procedure to the structural analyses by means of

direct strength assessment for ship less than 170 m in length. Unfortunately there is
no class procedure of direct strength assessment for ship less than 170 m in length.

The ship we analyzed is less than 170m and it has two cargo hold .Basically almost
all class society offer three cargo hold model. For these reasons it is impossible to
apply class rules to this ship for direct strength assessment.

Moreover, we developed a different procedure which is a combination of these two

methods and this method was applied to this ship as well. Method is that, we create
two cargo hold of the ship as well as finely meshing middle part and (half of the
cargo hold one, bulkhead, and half of the cargo hold two) and coarsely meshing the
other part.

As I mentioned above, this is just start of a work. My main aim is to develop an

analyses procedure to small ships.

29
30
REFERENCES

[1] Yong B., 2006. Marine Structural Design, Composite Structures, Article in
Press
[2] Ansys Inc., 2005. Release 10 Documentation for Ansys, Ansys Inc.
[3] Huges Owen F., 1988. Ship Structural Design, The Society of Naval Architect
and Marine Engineer
[4] Yao Tetsuya., 2003. Hull Girder Strength, Marine Structures, 16(2003), 1-13
[5] Sun Hai-Hong, Spencer Jack., 2005. Buckling Strength Assessment of
Corrugated Panels in Ofshore Structure, Marine Structures, 18 (2005)
548-565.
[6] ParkW-S., Fuiji D., Fujitani D., 1995. Finite element Analysis of
Discontinuous Thin Walled Beams Considering Nonuniform Shear
Warping Deformation, Computer & Structure, Vol 65 No 1 pp17-27,
1997
[7] Iijima K., Sheigemi T., Miyake R., KumanoA., 2004. A Practical Method for
Torsional Strength Assessment of Container Ship Structure, Marine
Structures, 17(2004), 355-384
[8] Cui W., Wang Y. Pederse P.T., 2002. Strength of Ship Plate Under
Combined Loading, Marine Structures, 17(2002), 75-97
[9] Servis D., Voudouris G., SamuelidisM., Papanikolaus A., 2003. Finite
Element Modelling and Strength analysis of Hold No 1 of Bulk
Carrier, Marine Structures, 18 (2005), 548-565
[10] Jia J., Ulfvarson A., 1994. A Parametric Study for Structural Behaviour of
Lightweight Deck., Engineering Structure, 26 (2004)963-977
[11] Umezaki K., Taguchi Y. Mori M., 1972. Evaluation of Hull Girder Stress on
The Open Ship, JSNA of Japan Vol 132 1972
[12] Paik K.J., Lee M. J., Kim W.C., 1999. Time Variant Ultimate Longitudional
Strength Corroded Bulk Carriers, Marine Structures, 16 (2003), 567-
600.
[13] Frickle W.,Cui W., Kihl D., 2002. Comperative Fatigue Strength Assessment
of a Structural Detail in a Container Ship Using Various Approaches
of Classification Societies, Marine Structures, 15 (2002), 1-13

31
[14] R.J. Astley, 1992. Finite Element in Solid andStructure, Chapman&Hall,
Cambridge
[15] B.V Class Rules
[16] E. Madenci, I.Guven, 2006. The Finite Element Method and Application in
Engineering Using ANSYS, Springer Science + Bussiness Media LLC
[17] E.G. Thompson, 2005. Introduction to Finite Element Method, Jhon Willey &
Sons Inc. New York
[18] J.C Card, 1996. Guidance for Eveluation of Finite Element and Result, Ship
Structural Commite, New York
[19] Prof. Dr. S. Tezcan, 1970. Çubuk Sistemlerin Elektronik Hesap Makinaları ile
Çözümü, Arı Kitabevi Matbası, İstanbul.
[20] J.P. Comstock, 1967 Principle of Naval Architect, The Society of Naval
Architecture and Marine Engineer, New York
[21] S. Moaveni, 1999. Finite Element Analyses Thearo and Application with
ANSYS, Prentice Hall, New Jersey
[22] Y.Nakasone, S.Yoshimoto, T.A.Stolarsici, 2006. Engineering Analysis with
ANSYS Software, Elsevier, Butterworth-Heinemann, Oxford
[23] T. Lamb, 2003. Ship Design And Construction. The Society of Naval
Architecture and Marine Engineer, New York

32
APPENDICES

APPENDIX A:

Figure A.1 : Shell thickness.

Figure A.2 : Stress points.

33
APPENDIX B

Figure B.1 : Sea pressure loading values of load case A (N/mm2).

Figure B.2 : Sea pressure loading values of load case B (N/mm2).

34
 Figure B.3 : Sea pressure loading values of load case C (N/mm2).

Figure B.4 : Internal pressure loading values of load case A (N/mm2).

35
Figure B.5 : Internal pressure loading values of load case B (N/mm2).

Figure B.6 : Internal pressure loading values of load case C (N/mm2).

36
 moment

100

50

0
32

-50

-100
bm(kNm/1000)

-150 moment

-200

-250

-300

-350

-400
frame

Figure B.7 : Vertical bending moment distribution along the model at loading
condition A.

moment

0
32
36
40
44
48
52
56
60
64
68
72
76
80
84
88
92
96

0
4
8
2
6
0
4
8
2
6
0
4
8
2
6
0
10
10
10
11
11
12
12
12
13
13
14
14
14
15
15
16
-50

-100

-150

-200
bm(kNm/1000)

-250 moment

-300

-350

-400

-450

-500
frame

Figure B.8 : Vertical bending moment distribution along the model at loading
condition B.

37
APPENDIX C

Figure C.1 : Von misses stress result at load case A (MPa).

Figure C.2 : Y direction displacement at load case A (mm).

38
Figure C.3 : YZ direction share stress result at side girders at load case A (MPa).

Figure C.4 : XY direction share stress result at side girders at load case A (MPa).

39
 Figure C.5 : Von misses stress result at side girders at load case A (MPa).

Figure C.6 : Von misses stress result at double bottom at load case A (MPa).

40
 Figure C.7 : YZ direction share stress result at double bottom at load case A
(MPa).

Figure C.8 : XZ direction share stress result at side girders at load case A (MPa).

41
Figure C.9 : Von misses stress result at load case B (MPa).

Figure C.10 : Y direction displacement at load case B (mm).

42
Figure C.11 : YZ direction share stress result at side girders at load case B (MPa).

Figure C.12 : XY direction share stress result at side girders at load case B (MPa).

43
Figure C.13 : Von misses stress result at side girders at load case B (MPa).

Figure C.14 : Von misses stress result at load case A (MPa).

44
Figure C.15 : YZ direction share stress result at double bottom at load case B (MPa).

Figure C.16 : Von misses stress result at double bottom at load case B (MPa).

45
Figure C.17 : Von misses stress result at load case C (MPa).

Figure C.18 : Y direction displacement at load case C (mm).

46
 Figure C.19 : Von misses stress result at side girders at load case C (MPa).

Figure C.20 : XY direction share stress result at side girders at load case C (MPa).

47
 Figure C.21 : Von misses stress result at double bottom at load case C (MPa).

Figure C.22 : XY direction share stress result at side girders at load case C (MPa).

48
CURRICULUM VITAE

Candidate’s full name : Barış DEDETAŞ

Place and date of birth : Istanbul 1979

Permanent Address : Çamlık Mah. Yeni Sok Bahar Evleri C/25 Çekmeköy
/Istanbul

Universities and
Colleges attended : Istanbul Technical University

49