See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/337360972

Detailed Finite Element Analysis of 180 M Deck Cargo / Launch Barge B42

Conference Paper · November 2019

DOI: 10.2118/197548-MS

CITATIONS READS
0 2,563

4 authors, including:

Fikri Bashar Yalchiner

National Petroleum Construction Company
2 PUBLICATIONS 12 CITATIONS

SEE PROFILE

All content following this page was uploaded by Fikri Bashar Yalchiner on 23 December 2020.

The user has requested enhancement of the downloaded file.

SPE-197548-MS

Detailed Finite Element Analysis of 180 M Deck Cargo / Launch Barge B42

Fikri Bashar Yalchiner, Rajender Agrawal, Faris Kamal, and Oussama Takieddine, National Petroleum Construction
Company, Abu-Dhabi

Copyright 2019, Society of Petroleum Engineers

This paper was prepared for presentation at the Abu Dhabi International Petroleum Exhibition & Conference held in Abu Dhabi, UAE, 11-14 November 2019.

This paper was selected for presentation by an SPE program committee following review of information contained in an abstract submitted by the author(s). Contents
of the paper have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect
any position of the Society of Petroleum Engineers, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written
consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may
not be copied. The abstract must contain conspicuous acknowledgment of SPE copyright.

Abstract
In this study, a detailed finite element analysis methodology of a 180 m deck cargo/launch barge B42 for
32,000 MT topside load is described and results are presented. Using Finite Element Method (FEM) for large
welded steel structures such as a deck cargo/launch barge has been quite challenging in past because the
engineer had to observe the balance between solution accuracy and numerical efficiency. However, recent
improvements in solver algorithms in FEM packages and increasing core processor numbers in computers
enable engineers to include a lot more details to their FE model so that accurate stiffness and mass of
the structure can be simulated. In this study, the entire barge is modelled in ANSYS Software using shell
elements including longitudinal beams, stiffeners, flanges, base plates, skid tracks and outriggers.
Two most critical load cases were analyzed; the first is the maximum hogging condition which occurs
in the topside loadout sequence where the topside is entirely on barge deck towards stern. The second
load being the maximum sagging case where the topside is at the final location on the deck of barge for
transportation.
Results of the detailed FE analysis confirmed the structural integrity of the barge showing all the stresses
and displacements are below allowable limits for all load cases. Two main advantages were observed
through detailed modelling of the entire barge. Firstly, a faster preprocessing time is as compared to shell-
beam models. Secondly, increasing the mesh density in critical locations in global FE model will be
equivalent to a sub-model in this case. Thus, eliminating the need for analyzing any detail separately.

Introduction
Cargo barges for float over decks must have sufficient structural strength to resist huge loads encountered
during loadout, transportation & float over. These barges require a rapid ballasting system in order to
compensate for skid shoe load during loadout & also for disengaging substructure from topside quickly
during float over. Ballast pumps for this purpose are usually housed in a separate pump room located within
the body of a barge. This results in discontinuity of barge longitudinal bulkheads which are the main load
carrying elements. Detailed structural adequacy check of barge is required in order to ensure the safety of
operation.
2 SPE-197548-MS

For this purpose, several load combinations were studied for loadout, transportation and floatover and the
worst scenario of loading in the hogging and sagging case was selected for detailed finite element analysis.
Due to schedule constraints the design of the barge was originally carried out by coarse and approximate
modeling and based on certain assumptions. However, prior to actual execution of the job, the barge was
modelled in detail and actual loading scenarios were investigated. This paper investigates the results of the
performed analyses. The topside was successfully installed subsequently in Arabian Gulf.
Main particulars of the barge are shown in Table 1.

Table 1—Main particulars of Barge B42

Length (Moulded) 180 m

Breadth (Moulded) 42 m

Breadth (Moulded) 72.04 m

Depth (Moulded) 11 m

Draft (loadline) 8.2 m

Sponsons

Length 50 m

Breadth 15.02

Depth 8.45

Material grade and extent of the barge are shown in Table 2.

Table 2—Material Grade and Extend of the Barge B42

Material Yield Stregth (MPa)

Grade AH36 355 Fr 0- Fr 242 +100 mm

Grade A 235 Elsewhere

Typical midship section of the barge is shown in Figure 1.

Figure 1—Typical Midship Section of the Barge

3D CAD model of the barge, which is used for the finite element model, including the skid tracks, base
plates and outriggers are shown in Figure 2.
SPE-197548-MS 3

Figure 2—3D CAD model of the barge and detail view of skid tracks, base plates and outriggers.

The barge is composed of 36 ballast tanks as shown in Figure 3.

Figure 3—Ballast tank Plan of B42 Barge

There are four pumps available on-board, two 7000 m3/h capacity one in the aft pump room and two
6000 m3/h capacity one in the forward pump room.
The topside has eight legs sitting on a skid shoe resting on skid track and loaded out from fabrication
yard to the barge via skidding. The position of the legs on skid shoes when 100 % of the topside is on the
barge (Stage 8) stern is shown in Figure 4.
4 SPE-197548-MS

Figure 4—The Position of the Topside on the Barge during Loadout Stage 8 (Hogging Case)

The final position of the topside on the barge during transportation superimposed with 5.00 m sagging
wave is shown in the Figure 5.

Figure 5—The Position of the Topside on the Barge during Transportation (Sagging Case)

Calculation Methodology
3 D CAD Model
3D CAD model preparation is the first and most cruicial stage in FE model preparation. When modelling a
very detailed, prominent structure like a cargo barge, it is essential to choose an efficient modelling software.
ANSYS Design Modeler was selected to prepare the 3D surface model of the barge using 2D drawings.
First, full 3D CAD model was completed, and then it was divided into nine different geometries along
the barge, meshed separately and then merged to create the full FE model (Figure 6 and 7) because of the
amount of details cannot be imported in ANSYS Mechanical as one geometry at one time.
SPE-197548-MS 5

Figure 6—Project Schematic from ANSYS Workbench showing the combination of FE models into one model
6 SPE-197548-MS

Figure 7—Transverse section details of the barge from 3D CAD model

Finite Element Model

In thin-walled metal structures such as barges, in order to save computational time shell elements are mainly
used to create finite element model. Shell elements are commonly used to model structures where two
dimensions are much larger than the third one (thickness) and when the change of the analyzed stress values
through thickness direction can be neglected.
The FE model of the barge consists of 5,430,662 nodes and 5,586,866 SHEL181 (four noded) elements
(Figure 8 and 9). The FE model is generated with 130 mm element size using SHELL181 which is suitable
for analyzing thin to moderately-thick shell structures. It is a four-noded (linear) element with six degrees
of freedom at each node: translations in the x, y, and z directions, and rotations about the x, y, and z axes [1].

Figure 8—FE model of the barge -Typical transverse frame

SPE-197548-MS 7

Figure 9—FE model of the barge - Skid track and base plates on barge deck (Stern side)

Longitudinal beams, brackets, and flanges are all modelled as surfaces and meshed as shell elements.
The main reason for this approach was to save time in preprocessing because connecting beam elements to
shell elements for the entire barge is a lot more time consuming process.

Load Case Combinations

Global structural analysis of a barge requires all the downward and upward forces be in equilibrium. The
donward forces can be listed as topside weight, barge's hull weight and additional structures' weights (skid
tracks, skid shoes & sponsons etc.) and the upward forces are buoyancy forces.
All loadings considered in the analyses are quasi-static. Quasi-static load means the load is applied so
slowly that the structure also deforms very slowly and as a result, the inertial and transient forces can be
neglected.
As a wave passes by a barge, the worst hogging moment will occur when the middle section is on the crest
of a wave and the bow and stern are in the troughs. This moment makes the barge to concave downward
creating tensile stresses on barge deck and compressive stresses on barge bottom plate.
The worst sagging moment will happen when the bow and stern of the barge are on two wave crests,
and the middle section in the trough between. This moment makes the barge to concave upward, creating
compressive stresses on barge deck and tensile stresses on barge bottom plate.
The global analysis has been performed for following two most critical load case combinations:
Maximum hogging condition during Loadout Stage 8. This occurs when entire topside has positioned
on the barge deck. (Figure 10)

Figure 10—Loadout Stage 8 - Maximum Hogging Moment (100% Row H Skid Shoe on Barge Aft at 1.80 m Tide)
8 SPE-197548-MS

Maximum sagging condition during Transportation of Topside including superimposition of 5.00 m

sagging wave. (Figure 11)

Figure 11—Transportation - Maximum Sagging Moment (5.00 m Sagging Trochoidal Wave)

For both load combinations static reaction forces of topside legs used in analyses are shown in Figure
12 and Table 3.

Figure 12—Static Reaction Forces of Topside Legs

Table 3—Static reaction forces of topside legs

REACTION (MT)

GRID MARKING ROW-3 ROW-5 TOTAL

ROW-B 4,600.7 3,456.0 8,056.6

ROW-D 4,587.6 3,590.3 8,177.9

ROW-F 4,851.0 3,632.3 8,483.3

ROW-H 4,487.4 2,799.7 7,287.1

TOTAL = 32,004.9 MT

Skid Shoe Reaction Forces on Skid Tracks

Skid shoe reaction forces on skid tracks are shown on Figure 13.
SPE-197548-MS 9

Figure 13—Static skid shoe reaction forces applied on skid tracks (sagging case)

Ballast Tank Loads

Ballast tank loads were applied according to ballast plan for both load case combinations. All wet surfaces
of the tanks are loaded with hydrostatic pressure. The most critical case occurs when a ballast tank is empty,
while the surrounding tanks are fully ballasted. This situation observed in maximum sagging case where the
ballast tanks under the topside were empty and neighbouring tanks were fully ballasted. A similar situation
is shown in Figure 14.

Figure 14—Ballast tank load (in MPa) for BWT1A.S – Sagging case
(Some plates are hidden to show the hydrostatic pressure distribution.)

Wave and Hydrostatic Loads on Barge

Waves at sea tend to be in trochoidal shape, rather than simple sine waves. Trochoidal waves have the
characteristic that the crests are steep and the troughs are more rounded. The trochoidal wave theory has
been adopted for specific standard calculations e.g. that for longitudinal strength. It is observed that the
approach appears to reflect many actual ocean wave phenomena accurately. Figure 15 shows buoyancy
force application for sagging case.
10 SPE-197548-MS

Figure 15—5.00 meter sagging trochoidal wave applied on barge hull. (Sagging case)

Sponsons have been accounted for by applying a distributed load to each side at the correct longitudinal
position, equivalent to the resultant load/moment which represents the summation of the forces at the centre
of gravity of each sponson. Sponson net buoyancy forces are shown in Figure 16.

Figure 16—Net buoyancy forces applied on sponson to barge connections (Sagging case)

Fixities
For a floating structure, it is crucial to achieve equilibrium while performing a finite element analysis. An
imbalanced model causes free body motion and hence unrealistic results. In order to restrain the model
a "321" constraint system was used avoid introducing an unrealistic moment constraint. This provides
the minimum amount of restraint while still allowing the model to run without errors due to unrestrained
degrees of freedom. The positions of the constraints are shown in Figure17 and 18. This consists of three
displacements, displacement A fixes the model in all translational degrees of freedom (XYZ), displacement
B fixes the model only in the vertical direction (Y) and displacement C fixes the model vertically and
transversely (YZ). This applies the minimum degrees of freedom consistent with a stable model. [5]
SPE-197548-MS 11

Figure 17—321 fixity on barge.

Figure 18—321 fixity - Applied on FE model.

The purpose of the constraints is to pick up the residual forces that occur due to slight imbalances between
the mass, applied loads and buoyancy forces.

Stress Acceptance Criteria

The von-Mises stress (obtained from the finite element stress components), should not exceed a certain
portion of the material's yield strength. It is generally expected that finer finite element mesh induces higher
resultant stress from a linear elastic analysis. However, the increase in stress is not just a function of finite
element mesh size. Field stresses are indicative of stress severity sufficiently away from structural details
such as hopper knuckles, openings and bracket toes. The recommended basic mesh size for capturing field
stresses is one longitudinal spacing. Element stresses directly obtained from 3D finite element models of
one longitudinal spacing can be considered as field stresses. For main supporting members, field stresses are
primarily due to primary hull girder deformation and secondary bending between watertight boundaries. In
practice, mesh size up to 1/3 longitudinal spacing is often used to calculate field stresses in main supporting
members. [2]
The shear stress allowables are taken from [6] and the allowable equivalent stress is taken from 4.7.1 of
[3] with the Sm, Strength Reduction Factor value taken from 2.5.3 Table 3 of [4].
Allowable stress values for Grade A and Grade AH 36 are shown in Table 4.

Table 4—Allowable stress values

Material Yield Strength Allowable Stress (MPa)

(MPa)
Shear Stress Equivalent Stress

Grade AH36 355 153 284

Grade A 235 110 188

12 SPE-197548-MS

Results
Only sagging case results are presented in this paper for discussion.

Results for Sagging Case

Reaction Forces at Fixities

Ideally, in a floating structure like a cargo barge, all boundary conditions should have zero reaction to ensure
that the load applied downwards is equal to the load applied upwards, as in reality there is no such thing
happening. The following are the reaction forces for each fixity (Table 5.). Total reaction forces at fixities
are significantly small, less than 1 % of the cargo barge's weight.

Table 5—Reaction forces at fixities.

Reaction Y Direction (N) Y Direction (MT)

1 548134.0489 55.87503047

2 -3319766.511 -338.4063722

3 2841712.914 289.6751187

Total 70080.4519 7.143776952

Comparison of Results from ANSYS and GHS (General HydroStatics) for Shear Force and
Bending Moments
For the maximum sagging moment case, there is a good correlation for shear force and bending moment
values obtained from GHS and ANSYS. GHS calculates shear forces and bending moments based on simple
beam theory assuming the cross section is constant (midship) throughout its length, while in ANSYS all
structural details of the barge are included in FE model. Figure 19 shows the shear force distribution from
GHS and ANSYS.

Figure 19—Shear Force Distribution for Maximum Sagging Case (GHS vs ANSYS)
SPE-197548-MS 13

For the maximum sagging moment case Figure 20 shows the Bending Moment distribution from GHS
and ANSYS.

Figure 20—Bending Moment Force Distribution for Maximum Sagging Case (GHS vs ANSYS)

Stress plots of longitudinal bulkheads show that the stresses are below allowables . Von Mises Stress
distribution for center line longitudinal BHD is shown in Figure 21.

Figure 21—Von Mises Stress Distribution for Center Line Bulkhead

YZ Shear Stress distribution for center line longitudinal bulkhead is shown in Figure 22.
14 SPE-197548-MS

Figure 22—YZ Shear Stress Distribution for Center Line Bulkhead

Stress plots of transverse bulkheads show that the stresses are below allowables. Von Mises Stress
distribution for transverse bulkheads is shown in Figure 23.

Figure 23—Von Mises Stress Distribution for All Transverse Bulkheads

Von Mises Stress distribution for transverse bulkhead under maximum topside static reaction force of
47588.3 kN is shown in Figure 24.

Figure 24—Von Mises Stress Distribution for Transverse Bulkhead

Under the Maximum Topside static Reaction Force of 47588.3 kN.

Total deformation distribution of entire barge for the sagging case is shown in Figure 25.
SPE-197548-MS 15

Figure 25—Total deformation distribution of entire barge. (Sagging case)

Conclusions
From this study, the following conclusions can be reached are as below
1. Detailed finite element analysis of deck cargo / launch barge during loadout (maximum hogging) and
transportation (maximum sagging) of a 32,000 MT has been presented.
2. Preparation of the 3D CAD model from the drawings is the most important step for FE model
preparation. It is one of the most time consuming process other than FE model generation. When it
comes to modeling a complicated, massive structure like a cargo barge it is vital to choose the correct
modelling software. ANSYS Design Modeler was chosen for modeling in this difficult task due to
its unique features such as, having a tree structure which makes going back in modeling process and
making changes easily and proceeding to model.
3. Developing a detailed FE model by using combination of nine seperate FE models in ANSYS
enables engineers to easily deal with simulations of large structures. Preparing separate FE models
simultaneously by different engineers can reduce the total modelling time drastically.
4. Having a fully balanced and detailed FE model is crucial to obtain accurate stress and deformation
results when analyzing a floating large structure like a cargo barge.
5. The ANSYS and GHS results showed good correlation of shear forces and bending moments for cargo
barge sections. This validates the global behavior of the shell FE model of the cargo barge.
6. 321 Fixity approach adequately utilized to constrain the FE model while not creating any artificial
stresses. The reaction forces found on fixities are smaller than 1% of the barge weight.
7. Stress values around large openings in longitudinal bulkheads at aft the and forward pump room areas
are below allowable limits.
16 SPE-197548-MS

8. Deck leg forces are accurately transferred to barge transverse and longitudinal bulkheads via base
plates, skid tracks and outriggers that are included in FE model.
9. The deflection of the barge is reasonable with a maximum deflection of 306 mm in the sagging
condition considering 180 m barge length and 11 m depth. This value is 15% lesser than the most
stringent beam deflection criteria of L/500 = 180000/500 = 360 mm.

NOMENCLATURE
ABS American Bureau of Shipping
DNV Det Norske Veritas
CAD Computer Aided Design
FEM Finite Element Method
ANSYS General Purpose Finite Element Software
GHS General HydroStatics Software
BHD Bulkhead
Sm Strength Reduction Factor

References
1. ANSYS Mechanical APDL Element Reference
2. The ABS Guide for Dynamic Loading Approach’ for FPSO Installations 2010.
3. ABS Guidance Note, Safehull Finite Element Analysis of Hull Structures, 2004
4. ABS Guide For Application Of Higher-Strength Hull Structural Thick Steel Plates In Container
Carriers, 2009 (2014)
5. DNVGL-CG-0127 ‘Class Guideline - Finite Element Analysis’, 2.5 Boundary Conditions
6. ABS Steel Vessel Rules

View publication stats