A STRUCTURAL DESIGN APPROACH FOR CONTROLLING WELDING DISTORTION AT THE UPPER DECK OF A HULL STRUCTURE IN THE ERECTION STAGE

A STRUCTURAL DESIGN APPROACH FOR CONTROLLING WELDING DISTORTION AT THE UPPER DECK OF A HULL STRUCTURE IN THE ERECTION STAGE S.-B. Shin, D.-J. Lee and J.-G. Youn

ABSTRACT

An effective design method for controlling the transitional welding distortion at the aft upper deck occurring in the erection stage of the hull structure has been established using FEA and measured distortion database. Variation of distortion at the aft upper deck was measured according to the progress of the manufacturing process. This was done in order to clarify the 2nd distortion of the upper deck, defined as the increment of welding distortion developed after the assembly stage. The amount and distribution of the 2nd load causing the 2nd distortion were identified with a non-linear buckling FEA. The distribution of the welding distortion and the residual stress were assumed to be initial imperfections, which were identified with STEM (simplified thermoelastic method) based on inherent strain. Based on these results, the modification of structural design parameters such as plate thickness and the attachment of carling were made, in order to prevent the excessive 2nd distortion which requires costly flame correction work in the erection stage. IIW-Thesaurus keywords: Arc welding; Buckling; Distortion; Finite element analysis; Liquefied gases; Ships; Shipbuilding; Stress analysis; Structures.

51

1

Introduction

Excessive welding distortion introduced during shipbuilding has required costly correction work such as line heating and spot heating. So, extensive research works on the control of welding distortion for hull structures have been performed, in order to enhance the productivity of shipbuilding. However, the excessive welding distortion of hull structures has been continuously reported. This may be attributed to the fact that most researches [1~6] have focused on reducing the analysis cost in the evaluation of the welding distortion of hull structures, based on the inherent strain methods. Generally, commercial vessels are constructed through a series of manufacturing processes: sub-assembly of unit components, assembly of unit block and erection. Although the welding distortion is corrected after the assembly stage, the excessive distortion at the upper deck develops in the erection stage. This also requires costly correction work in order to control the distortion to below the allowable distortion specified before launching the ship. This is due to the effect of the manufacturing process after the assembly stage. That is, after the assembly stage of the unit block, the blocks are exposed to various loads induced by post-

manufacturing processes, including lifting and turn-over of the assembled block, various welding processes for the fittings, pre-erection and erection in a dock for shipbuilding. Therefore, in order to minimize the distortion of hull structures in the erection stage, the principal factors controlling welding distortion in the whole manufacturing process for shipbuilding should be identified. The purpose of this study is to establish a new design method for the control of excessive welding distortion of the aft upper deck in the erection stage. For this, the transitional behaviour of welding distortion of the aft upper deck was measured with the progress of the manufacturing process. This provides the amount of the 2nd distortion of the upper deck. (In this study, “2nd distortion” is defined as the increment of welding distortion developed during the erection stage.) With the amount of the 2nd distortion, the 2nd load corresponding to the 2nd distortion was determined by a non-linear buckling FE analysis. Here, the distribution of welding distortion and residual stress in the assembly stage were assumed to be initial imperfections. The map for the distribution of the 2nd load at the upper deck was established. Based on these results, the modification of structural design was proposed in order to avoid the excessive 2nd distortion of the upper deck in the erection stage.

Doc. IIW-2246, recommended for publication by Commission XV “Design, Analysis and Fabrication of Welded Structures.”

N° 03 04 2012 Vol. 56 WELDING IN THE WORLD

Peer-reviewed Section

A STRUCTURAL DESIGN APPROACH FOR CONTROLLING WELDING DISTORTION AT THE UPPER DECK OF A HULL STRUCTURE IN THE ERECTION STAGE

2

Design and analysis method

2.1 STEM for welding distortion In this study, STEM (simplified thermoelastic method) was used for evaluating the distribution of welding distortion of the upper deck in the assembly stage, which was one of the initial imperfections for evaluating the 2nd load. In STEM, welding distortion was evaluated by elastic FEA solver and thermal loads, which were determined with the inherent strains corresponding to the angular distortion, the transverse and the longitudinal shrinkage of the weldment. All variables used for STEM were determined on the basis of the results of the simple welds obtained using a conventional FEA. This method is very similar to the linear elastic shrinkage volume method (LESVM) [7].

But, for the STEM, the welding distortion was evaluated using not the eight-node brick element, but the four-node shell element. Therefore, it is a more economical approach than the LESVM. Figure 1 shows variations of the angular distortion and shrinkage force at the simple fillet weldment, with respect to welding heat input (QO) and base plate thickness in-plane (t). These were obtained by the comprehensive FEA, dimensional analysis and experiment [8, 9]. Here, the dimension of test specimen and welding condition used for FEA and experiment are given in Table 1. As shown in Figure 1, the angular distortion and the transverse shrinkage and the longitudinal shrinkage force could be defined as the function of welding heat input (Qo) and the rigidity of the weldment, respectively. From the results, the inherent strains corresponding to the angular distortion and the transverse shrinkage were

52

a) Angular distortion

b) Transverse shrinkage

c) Longitudinal shrinkage force Figure 1 – Variation of welding distortion at the simple fillet weldment with welding heat input and rigidity of the weldment [8, 9]

A STRUCTURAL DESIGN APPROACH FOR CONTROLLING WELDING DISTORTION AT THE UPPER DECK OF A HULL STRUCTURE IN THE ERECTION STAGE

Table 1 – Dimension of simple fillet weld and welding conditions [7, 8] Welding condition

Stiffener

Base plate thickness [mm]

Type

Thickness [mm]

Height [mm]

Process

Heat input [kJ/cm]

12 ~ 25

Flat bar

12~25

250

FCA

13.5

calculated by using Equations (1) and (2). Here, δTop and δBot represent the amount of shrinkage at the top and bottom surfaces of the weldment, as shown in Figure 2.

GTop

GS 

G Bot

GS 

GA

(1)

2

GA

(2)

2

where

δTop is the transverse shrinkage at the top (welding) side [mm], δBot is the transverse shrinkage at the bottom side [mm], δA is the transverse shrinkage due to angular distortion [mm], δS is the transverse shrinkage [mm]. Here, the thermal loads for inherent strains were defined with temperature difference in the thickness direction and the linear thermal expansion coefficient by using Equations (3) and (4).

GTop G Bot

D'T1

D (TR  TTop )

Le

Le

D'T1

D (TR  TBot )

Le

Le

conditions are shown in Table 2. For EELM, the equivalent bending moment and shrinkage force corresponding to inherent strain were applied to the FE model instead of thermal load. As shown in Figure 4, the deformed profiles of the panel structure obtained by STEM and ELEM have good agreement with the measured results. Figure 5 shows the deformed profiles of a SA butt weldment, with the width of the base plate obtained by both STEM and by experiment. Here, the Y axis is the dimension-less variable for angular distortion, defined as the ratio of maximum angular distortion to the amount of angular distortion at each location of the butt weldment. The plate thickness is 12 mm. As shown in Figure 5, the deformed profile obtained by STEM is almost the same as the one obtained by measured results. It means that the effect of self-restraint, such as self-weight on the welding distortion of the weldment, can be easily predicted by STEM. Therefore, in this study, STEM was employed to determine the welding distortion of the aft upper deck in the assembly stage.

(3)

(4)

where

α is the Linear thermal expansion coefficient, Figure 2 – Inherent strain at the weldment

Le is the element length corresponding to the weldment [mm], TR is the initial reference temperature in Figure 3, TTop is the temperature at the top side of weldment, TBot is the temperature at the bottom side of weldment. As given in Equations (3) and (4), once TR was set to be molten temperature of weld metal, other variables, TTop, TBot and α could be defined by using the transverse shrinkage in the top and bottom sides of the weldment. Figure 3 shows the relationship between each variable used for STEM. Figure 4 shows the distribution of welding distortion obtained by STEM, EELM (Elastic Equivalent Load Method) and by experiment. The dimension of the panel structure used as the test specimen and the welding

Figure 3 – Temperature variables used for STEM

N° 03 04 2012 Vol. 56 WELDING IN THE WORLD

Peer-reviewed Section

53

A STRUCTURAL DESIGN APPROACH FOR CONTROLLING WELDING DISTORTION AT THE UPPER DECK OF A HULL STRUCTURE IN THE ERECTION STAGE

Figure 4 – Welding distortion of the panel welded structure consisting of fillet weldments obtained by experiment, STEM and EELM

Figure 5 – Welding distortion at the SA butt weldment obtained by experiment and STEM

Table 2 – Dimension of panel structure and welding conditions Stiffener Plate thickness [mm] 54

10

Welding condition

Dimension [mm]

Span [mm]

Process

Heat input [kJ/cm]

100 x 10 (Flat bar)

850

FCA

13.5

2.2 Evaluation of the 2nd distortion and the 2nd load In this study, the concept of 2nd distortion and 2nd load were introduced in order to explain the transitional behaviour of welding distortion at the aft upper deck in the erection stage. These were based on the concept responsible for the corrugation damage of the bottom shell plating with a transverse frame, as shown in Figure 6 [10, 11]. The previous results show that the major factors influencing the corrugation damage induced in the service life of a ship were identified as compressive stress, caused by the hull girder bending moment in the longitudinal direction of the ship, and water pressure. In addition, initial welding distortion and residual stress also had a great influence on corrugation damage. This is attributed to the fact that the initial imperfections decrease the buckling strength of the welded structure. Figure 7 shows the mechanism of corrugation damage in the hull structure. According to the

*1

increase in the amount of external stress applied to the deck structure, the amount of the distortion at the deck plate sharply increases. In addition, the deformed shape is also abruptly changed as the external applied stress exceeds the critical value. From the results, it can be inferred that there are 2nd loads inducing the transitional welding distortion at the aft upper deck after assembly stage. That is, if the upper deck has a sufficient resistance against the 2nd load produced during the manufacturing processes, 2nd distortion could be easily controlled. Therefore, in this study, 2nd distortion, defined as the increment of distortion from the assembly stage to the

HBM: Hull Girder Bending Moment.

Figure 6 – Bottom shell plating of a transverse frame [11]

Figure 7 – Mechanisms of corrugation damage [11]

A STRUCTURAL DESIGN APPROACH FOR CONTROLLING WELDING DISTORTION AT THE UPPER DECK OF A HULL STRUCTURE IN THE ERECTION STAGE

erection stage, is first determined from the database of welding distortion measured in each manufacturing process. In the next step, the amount and distribution of the 2nd load responsible for the 2nd distortion are identified by non-linear buckling FE analysis. In FEA, the welding distortion and the residual stress at the upper deck in the assembly stage are evaluated by STEM and assumed as initial imperfections for non-linear buckling FEA. Figure 8 shows an example of the external loads for the evaluation of the 2nd load corresponding to the 2nd distortion at the aft upper deck. Figure 9 shows the typical example of the relationships between the 2nd load and the 2nd distortion, obtained by non-linear buckling FEA.

Figure 8 – External loading conditions to identify the amount of the 2nd load

3 Design and analysis method for the 2 nd distortion

3.1 Analysis model and representative model Figure 10 shows the plan of the aft upper deck in a LPG carrier, selected as an analysis model in this study. As shown in Figure 10, the flame-straightened zone for the control of excessive distortion in the erection stage was found at the centre zone of the aft upper deck. In reference, the allowable distortion at the upper deck after the erection stage is set to be 6.0 mm. Figure 11 shows the amount of welding distortion around the flame-straightened zone at the upper deck, which was measured in the assembly and erection stages, respectively. As shown in Figure 11, except the welding distortion around the erection joint, most welding distortion of the upper deck measured in the assembly stage is less than 4.0 mm. However, the welding distortion in the erection stage

55

Figure 9 – Relationship between the 2nd load and the 2nd distortion at the upper deck

Figure 10 – Plan of the aft upper deck in a LPG carrier

N° 03 04 2012 Vol. 56 WELDING IN THE WORLD

Peer-reviewed Section

A STRUCTURAL DESIGN APPROACH FOR CONTROLLING WELDING DISTORTION AT THE UPPER DECK OF A HULL STRUCTURE IN THE ERECTION STAGE

Table 3 – Dimensions of main components in the aft upper deck Base plate thickness [mm]

Longi. stiffener [mm]

Girder and trans. stiffener [mm]

11 (STBD), 13 (Port)

150 x 90 x 9

600 x 12 + 150 x 20

sharply increased. This indicates that the transitional behaviour of welding distortion occurs in the erection stage. Figure 12 shows the configurations of simplified models used for evaluating the variation of welding distortion at the flame-straightened zone of the upper deck. The simplified models were constructed in consideration of

dimensional characteristics and manufacturing processes for the upper deck. Table 3 shows the dimensions of the main components of the simplified model. The welding conditions for each main component and manufacturing sequence in the assembly stage are given in Table 4 and Table 5, respectively.

56

(Unit: mm)

Figure 11 – Variation of welding distortion at the flame-straightened zone of the upper deck in a LPG carrier

Table 4 – Welding methods and welding heat input for each of the main components Welding method

Plate thickness [mm] 11

I Butt

SA 13

3VM a)

Fillet

a)

FCA

FCA

Heat input [kJ/cm]

Remark

33.2

Backing side

33.6

Finishing side

37.8

Backing side

39.9

Finishing side

11

53.5

13

66.3

11~13

3VM represents “V” groove weldment for erection welding with a 6 mm gap.

Total heat input

16.6

Primary member

15.3

Secondary member

9.4

Carling

A STRUCTURAL DESIGN APPROACH FOR CONTROLLING WELDING DISTORTION AT THE UPPER DECK OF A HULL STRUCTURE IN THE ERECTION STAGE

a) Port

b) STBD * Weld for PE represents the weldment for Pre-Erection

Figure 12 – Simplified model for the upper deck

57

a) Port (Unit: mm)

b) STBD

Figure 13 – Distributions of welding distortion in the simplified models after the assembly stage Table 5 – Manufacturing sequence for the simplified model No.

Description

1

SA I butt welding between main base plates

2

Fillet welding between base plate and secondary member

3

Fillet welding between base plate and primary member

4

FCA welding between primary and secondary member

5

Welding between Port and STBD blocks (3VM)

3.2 Initial imperfections The amount and distribution of welding distortion in the simplified model of the upper deck in the assembly stage were evaluated by using STEM. The welding heat input in each simplified model was adjusted to produce the same amount and distribution of welding distortion measured in the assembly stage. Figure 13 shows the distributions of welding distortion in the simplified models. In the assembly stage, the amount of maximum welding distortion in the simplified STBD model is slightly larger than that of

welding distortion in the simplified Port model, due to the difference in the main plate thickness as shown in Table 3. However, the global deformed shape of the simplified Port model with a plate thickness of 13 mm is more complex, compared with the STBD model which has carlings. Figure 14 shows the results of welding distortion in the simplified model obtained by STEM and measurement. The deformed shape and the amount of welding distortion obtained by STEM are almost identical to those obtained by measurement. It means that the distribution of welding distortion in the simplified models after the assembly stage can be used as an initial imperfection to evaluate the 2nd load by non-linear buckling FE analysis. The distribution of residual stress was evaluated by using Figure 1 c) and Table 4. Figures 15 and 16 show the distribution of transverse and longitudinal residual stresses in the simplified models. The width of the tensile residual stress zone of each simplified model was determined with the longitudinal shrinkage force and the assumption that the amount of residual stress is equal to the yield strength of the base plate. As reference, the residual stress induced by welding N° 03 04 2012 Vol. 56 WELDING IN THE WORLD

Peer-reviewed Section

A STRUCTURAL DESIGN APPROACH FOR CONTROLLING WELDING DISTORTION AT THE UPPER DECK OF A HULL STRUCTURE IN THE ERECTION STAGE

(Unit: mm)

Figure 14 – Distributions of welding distortion in the simplified models in the assembly stage

58

a) Longitudinal direction (Unit: MPa)

b) Transverse direction

Figure 15 – Distributions of residual stress in the simplified Port model after the assembly stage

a) Longitudinal direction

b) Transverse direction (Unit: MPa)

Figure 16 – Distributions of residual stress in the simplified STBD model

A STRUCTURAL DESIGN APPROACH FOR CONTROLLING WELDING DISTORTION AT THE UPPER DECK OF A HULL STRUCTURE IN THE ERECTION STAGE

in the assembly stage was applied to the initial stress at the element corresponding to the tensile residual stress zone.

responsible for the 2nd distortion, the external load was applied to the simplified model having initial imperfections shown in Figures 13 to 16.

3.3 Evaluation of the 2nd load

Figure 18 shows the distribution of the 2nd distortion in the simplified model with an external load of 0.72 kN/mm for the Port model and 0.85 kN/mm for the STBD model. As shown in Figure 18, the deformed shape in both models under the specified external load was a wave type. Figure 19 shows the relationship between the 2nd load and the 2nd distortion at “A” and “B” in each model. “A” and “B” in the models are the location of the maximum welding distortion in way of fillet and butt weldment in each model respectively. As shown in Figure 19, the amount of

Figure 17 shows the amount of the 2nd distortion at the flame-straightened zone shown in Figure 10. The amount of the 2nd distortion was determined as the difference of welding distortion between the assembly stage and the erection stage. As shown in Figure 17, the excessive 2nd distortion was found where the maximum welding distortion occurred in the assembly stage. For the evaluation of the amount and the distribution of the 2nd load

59

(Unit: mm)

Figure 17 – Distribution of the 2nd distortion at the flame-straightened zone

a) Port (Unit: mm)

b) STBD

Figure 18 – Contours of the 2nd distortion distribution in the simplified models with external load

N° 03 04 2012 Vol. 56 WELDING IN THE WORLD

Peer-reviewed Section

A STRUCTURAL DESIGN APPROACH FOR CONTROLLING WELDING DISTORTION AT THE UPPER DECK OF A HULL STRUCTURE IN THE ERECTION STAGE

a) Port

b) STBD

Figure 19 – Relationship between the 2nd load and the 2nd distortion at “A” and “B” in each model

60

the 2nd load increased with an increase in 2nd distortion. For the Port model shown in Figure 19 a), the amount of 2nd distortion at “A” was larger than that at “B”. However, for the STBD model, the amount of 2nd distortion at each location (“A” and “B”) was almost similar. This is mainly due to the difference of initial welding distortion at “A” and “B” in each model. From the results, the amounts of 2nd load at “A” and “B” of each model were identified as given in Table 6.

upper deck was performed. The main plate thickness in the STBD of the upper deck was changed from 11 mm to 13 mm. The carling (75 mm × 10 mmt) in the Port of the upper deck was attached to the main plate as shown in Figure 21. Figure 22 shows the behaviour of the 2nd distortion of the simplified models before and after the design change. As shown in Figure 22 a), in “A” of the simplified Port model under a 2nd load of 0.89 kN/mm, the amount of the 2nd distortion was reduced from -7.5 mm to 2 mm after the attachment of the carling. For the STBD simplified model with a 2nd load of 1.2 kN/mm, the 2nd distortion of 10.0 mm was also reduced to 1.8 mm. This result indicates that the design change proposed in this study could be applied as an alternative to avoid the flamestraightened work age at the upper deck in a LPG carrier after the assembly stage. Figure 23 shows the extent of the flame-straightened zone of the aft upper deck measured before and after the design change. As shown in Figure 23, the flame-straightened zone in the erection stage was substantially reduced after the design modification. In addition, the flame-straightened zone of the aft upper deck which occurred after the design change was due to the excessive welding distortion produced in the assembly stage. That is, the 2nd distortion in the erection stage was almost perfectly controlled by the design modification proposed in this study. Based on the results, it can be concluded that the distortion control method based on the 2nd load and the 2nd distortion can be used as one of the effective solution methods to prevent the excessive distortion in the erection stage.

3.4 Map of the 2nd load In order to identify the amount and the distribution of the 2nd load at the aft upper deck of LPG carrier, the relationships between 2nd load and 2nd distortion in other regions of the upper deck were investigated with the additional simplified models, which were constructed in consideration of the characteristics of design and manufacturing process. Figure 20 shows the distribution of the 2nd load at the upper deck of a LPG carrier. As shown in Figure 20, the amount of the 2nd load was relatively very low with the exception of the flame-straightened zone of the aft upper deck in the erection stage. This result indicates that the excessive 2nd distortion found at the upper deck of a LPG carrier could be easily controlled through modifying the partial structural design of the upper deck.

3.5 Design for the control of the 2nd distortion In order to prevent the excessive 2nd distortion at the upper deck of a LPG carrier, the design change of the

Table 6 – The amount of the 2nd load at “A” and “B” in each simplified model Port

STBD

Location nd

2 Load [kN]

A

B

A

B

0.89

0.09

1.206

0.44

A STRUCTURAL DESIGN APPROACH FOR CONTROLLING WELDING DISTORTION AT THE UPPER DECK OF A HULL STRUCTURE IN THE ERECTION STAGE

a) Port

(Unit: N/mm)

b) STBD

Figure 20 – The map of the 2nd distortion at the aft upper deck in a LPG carrier

61

a) Before design change

b) After design change

Figure 21 – Configurations of the simplified Port model before and after design change

a) A” in the Port model

b) “A” and “B” in the STBD model

Figure 22 – Variations of the 2nd distortion in the simplified models before and after design change

N° 03 04 2012 Vol. 56 WELDING IN THE WORLD

Peer-reviewed Section

A STRUCTURAL DESIGN APPROACH FOR CONTROLLING WELDING DISTORTION AT THE UPPER DECK OF A HULL STRUCTURE IN THE ERECTION STAGE

a) Before design change

62

b) After design change Figure 23 – Flame-straightened zone of the aft upper deck before and after the application of design change

4 Conclusions In order to control the excessive welding distortion and avoid the flame-straightened work for the hull structure in the erection stage, an effective design method based on 2nd load and 2nd distortion was proposed. The main results are summarized as follows. 1. The measured results of welding distortion during the manufacturing process show that the distribution and amount of welding distortion at the aft upper deck in a LPG carrier were changed in the erection stage. 2. The increment of welding distortion in the erection stage, the 2nd distortion strongly depends on the amount and distribution of initial welding distortion. It means that the allowable distortion criteria for accuracy management should be established in consideration of the difference of welding distortion in the

neighbouring zone as well as the amount of welding distortion. 3. In order to control the 2nd distortion in the erection stage, the design method for distortion control based on the 2nd load was established. Here, the amount of the 2nd load corresponding to the 2nd distortion was calculated by using non-linear buckling FE analysis with simplified model. The welding distortion and residual stress in the assembly stage were assumed to be initial imperfections. With the results, the map of the 2nd load at the aft upper deck plate in the erection stage was established. 4. Based on the results, the design modification for the aft upper deck was established with the change of the main plate thickness and the attachment of a carling. The validity of the design change was verified by the comparison with the extent of the flame-straightened zone before and after its application.

A STRUCTURAL DESIGN APPROACH FOR CONTROLLING WELDING DISTORTION AT THE UPPER DECK OF A HULL STRUCTURE IN THE ERECTION STAGE

References [1] Jang C.D. and Lee C.H.: A study on the prediction and control of welding deformation of ship hull blocks, Journal of the Society of Naval Architects of Korea, 2000, vol. 37, no. 2, pp. 127-136. [2] Jang C.D. and Seo S.I.: A simplified method to estimate longitudinal deformations of built-up beams due to welding and heating, Journal of Ship Research, 1995, vol. 39, no. 2, pp. 176-183. [3] Kim S.I. and Lee J.S.: Development of simple prediction method for welding distortion in fillet joints, Proceedings of the Annual Spring Meeting, The Society of Naval Architects of Korea, pp. 265-270.

[8] Shin S.B. and Yoon J.G.: Effects of internal and external restraint on welding distortion, Proceedings of the Annual Spring Meeting, The Society of Welding and Joining of Korea, 2001, pp. 165-168. [9] Shin S.B. and Lee D.J.: A study on the distortion behavior at the aft upper deck of the ship structure during the manufacturing process, Proceeding of SNAK, 2002, pp. 409-412. [10] Murray J.M.: Corrugation of bottom shell plating, Transactions of the Institute of Naval Architects, London, 1954, vol. 94, pp. 229-250. [11] Masubuchi K.: Analysis of welded structures: residual stresses, distortion, and their consequences, Pergamon Press, Oxford, 1980.

[4] Kim S.I., Cho Y.K., Lee H.W. and Lee J.S.: Prediction of welding deformation of panel block using simplified analysis method, 1996, Proceedings of the Annual Spring Meeting, The Society of Naval Architects of Korea, pp. 271-276. [5] Watanabe M. and Satoh K.: Effect of welding conditions on the shrinkage distortion in welded structures, Welding Journal, August 1961, vol. 40, pp. 377-384. [6] Nomoto T., Takechi S and Aoyama K.: Basic studies on accuracy management system based on estimation of welding deformation, Journal of the Society of Naval Architects of Japan, 1997, vol. 181, pp. 249-260. [7] Bachorski A., Painter M.J., Smailes A.J. and Wahab M.A.: Finite-element prediction of distortion during gas metal arc welding using the shrinkage volume approach, Journal of Materials Processing Technology, 1999, vol. 92-93, pp. 405-409.

63

About the authors Dr. Sang-Beom SHIN ([email protected]), Principal Researcher, Mr Dong-Ju LEE ([email protected]), Chief researcher and Dr. Joong-Geun YOUN (jgyoun@hhi. co.kr), Senior Vice President, are all with the Industrial Research Institute, Hyundai Heavy Industries. Co. Ltd., Ulsan (Korea).

N° 03 04 2012 Vol. 56 WELDING IN THE WORLD

Peer-reviewed Section