While carrying out welding work on the hull block joint, a plate jig, referred to as temporary piece, is generally used to correct misalignment and prevent welding deformation [
1]. However, because the deformations cannot be prevented and the weight increases if pieces become larger than a specific size, each shipyard uses a standardized piece, with specific shape and size.
Figure 1 shows the working area of a temporary piece in the block assembly stage,
Figure 2 shows the working area in the block erection stage, and
Figure 3 presents technical drawings for representative standardized temporary pieces and strong-back. In each shipyard, some hundred thousand temporary pieces are used in a year, according to its scale, and the production time required for piece installation, removal, grinding, and work completion also mounts to some hundred thousand hours. A survey found that around 38,000 temporary pieces are used for only one unit of a 50,000 deadweight tonnage (DWT) tanker [
2]. Furthermore, quality accuracy control of the hull block, by controlling deformations caused by welding, is very important for improving the productivity in the assembly and erection stages. Therefore, reduced usage of temporary piece, while maintaining a welding quality of hull block, is required.
Interest in the optimization of butt joint welding has generated many analytical and experimental studies [
3,
4,
5,
6]. Furthermore, the studies have been made to reduce residual stresses and deformations of fillet weld joints. Perić et al. carried out a numerical simulation for reducing longitudinal residual stress and deformation using a local preheating technique [
7]. The authors analyzed the correlation between residual stress and deformation according to preheating temperature and inter-pass time. Recently, efforts to precisely evaluate the residual stress and deformation of the butt-welded joint have been aided by finite element analysis, thanks to the development of increasingly more powerful computers. However, most of the research done has been limited to a range of simple welded joint model levels [
8,
9]. Finite element simulations of the welding deformation of large structures, such as aircraft fuselages and hull blocks, have the downside of generating a quite large deviation from the real welding deformations in spite of large calculation times, due to considering the nonlinearity of materials [
10]. It is known that the accuracy of the simulation result is largely relying on an appropriate consideration of the weld area affected by heat [
11,
12,
13]. Hernando et al. explained a laser beam welding (LBW) model to predict the geometry of the resulting joint when welding thin Inconel 718 plates used in the aerospace industry [
14]. Jang et al. and Park et al. carried out simulations of the welding deformation by the assembly sequence of reinforced plate of hull body, by combining the equivalent load method and finite element method based on the inherent strain [
15,
16]. Kim et al. proposed an equivalent strain method, based on inherent strain for curved double bottom ship block [
17]. Kang et al. modified the inherent strain method to include the friction stir welding process [
18]. This method expedited calculation time by simplifying a complicated thermal elasto-plastic analysis by using the inherent strain theory. However, this method has limitations in deciding the size and distribution area of inherent deformations. Furthermore, it requires an inconvenient process of obtaining a degree of restraint, each time, if external restraint condition is changed. Meanwhile, several studies have focused on the welding deformation itself. Deng et al. clarified the generation mechanism of angular distortion in fillet welded joints through numerical simulations and experiments [
19]. Adamczuk et al. developed a methodology to predict the angular distortion in multipass V-butt joint welding based on experimental and analysis results [
20]. They analyzed the behavior of angular distortion along the passes performed. Mochizuki and Okano investigated the effect of the welding process and heat input conditions on the angular distortion induced by bead-on-plate welding through a numerical approach [
21]. They developed the parameter of mechanical melting region on the plate thickness and found that it was the dominant factor for accurately quantifying angular distortions. Xie et al. revealed the mechanism of angular distortion in fusion welding and how the welding processes and factors influence the angular distortion [
22]. They established a theoretical model concerning the melting–solidification process and formulated an expression for angular distortion. Seong developed a systematic method to predict angular distortion in multilayer welding [
23]. He defined the relationship between heat input, bead cross-section, angular distortion, and thickness through the bead-on-plate welding experiment and constructed databases from them. Then he proposed a method of predicting angular distortion in multilayer welding and verified it through welding experiments on V-groove butt joints. Ryu and Kim et al. conducted studies on the butt-welding deformation using pieces [
24,
25]. Ryu only conducted experimental studies on the butt welding deformation at the level of unit specimens. Kim et al. simulated the butt-welding deformation of hull block joints, however they did not derive the optimal number of pieces and neither they simulate the butt-welding deformation of large hull blocks.
In this study, a thermal elasto-plastic analysis based on the finite element method was employed to examine the welding deformation. The necessary parameters were chosen based on the experimental results, to guarantee the accuracy of the results. This study was carried out with the objective of maintaining quality accuracy of hull block and reducing temporary piece usage. The restraint effect of the temporary piece on the deformation of base plates during joint welding was reviewed both qualitatively and quantitatively through a series of butt-welding experiments. The calculation and analysis results were compared and verified with the experimental data. Lastly, the welding deformation on the hull block joint, considering the temporary piece, was simulated by using a verified analysis method, and the results obtained were compared with the experimental results. Based on these results and by placing the temporary pieces at the right places, an attempt to achieve quality accuracy of the butt joint welding area in panel block and saving the temporary piece usage at the same time was made.