Keywords: welding, residual stress, plate girder, 3D finite element analysis.

In many industries welding plays an important role since it is the most frequently used joining method for metallic materials. In fusion welding such as arc welding and laser beam welding a concentrated heat source at a very high temperature is moved along the components to be joined. This makes the base material melt in a small area around the heat source. In this so called weld pool the molten material of the two components (with or without additional filler material) are mixed together and after cooling this solidified mixture provides the joint between the components.

During welding the microstructure of the material changes in the weld and in the heat affected zone, which results in changes in the material properties. Volume changes of the base material caused by phase transformation in the weld pool introduces additional stresses. Finally, as a result of the highly concentrated heat input an uneven temperature distribution develops in the welded components resulting in further stresses and deformations. The high tensile stresses near the weld, which can reach the yield limit, may reduce the strength of the structure and increases the chances of fatigue crack development. The compressive stresses, which maintain equilibrium with the tensile stresses in the unloaded structure, reduce the buckling strength of the structure. The deformations can be very significant and beside their unfavourable effects on the strength of the structure they also make the assembly of the structure more difficult [1,2].

Since residual stresses and deformations may seriously affect the behaviour of welded structures these imperfections have to be taken into consideration during their design. The current design practice uses semi-empirical methods based on extensive experimental studies as a design basis. The aim of the research described in this paper is to develop techniques for the simulation of the fabrication process of full-scale engineering structures in order to determine the imperfections of the structure and incorporate this knowledge alongside virtual experiments into the design procedure.

The main difficulties in ensuring an accurate numerical simulation of welding are the lack of experimental data, complex material behaviour, multiphysics nature of heat source and weld pool modelling as well as the large number of welds in a real structure. To overcome these difficulties and reduce the calculation time to an acceptable level the introduction of a number of simplifications is required. The most common techniques used to simplify the simulation are the use of decoupled thermal and stress analyses, use of 2D models instead of full 3D ones and use of simplified material modelling. The details of these difficulties and simplifications are discussed in the paper.

In order to develop an effective numerical tool to study complex welded structures the numerical methods need to be evaluated in the view of welding. The welding procedure can be divided into two well distinguishable parts both in time (welding and cooling) and in space (near the welds and the rest of the structure). The paper presents how these features of welding can help to determine an efficient numerical solution strategy.

Nowadays parallel and distributed computing is more widely available and parallelisation provides an excellent mean of reducing computational time. Although the use of parallel computers in welding simulation has been predicted by researchers [3,4], paper on parallelisation for welding simulation has not been found in the literature. One aim of the research is to study parallelisation techniques with respect to welding simulation. For parallel or distributed processing the numerical model needs to be broken up (partitioned) into as many pieces as the number of available processors in the computer. The efficiency of parallelisation is highly dependent on partitioning, during which the spatial features of welding can be exploited.

The paper presents two early examples: a plate girder with vertical stiffeners and a plate with bead-on-plate welding. These examples demonstrate very well the special features of welding and help us to determine where to focus during the further development.

1

K. Masubuchi, "Analysis of Welded Structures". Pergamon Press, 1980.

2

D. Radaj, "Heat Effects of Welding". Berlin Heidelberg: Springer-Verlag, 1992.

3

L.-E. Lindgren, "Finite element modeling and simulation of welding, part 3: Efficiency and integration," Journal of Thermal Stresses, vol. 24, pp. 305-334, 2001. doi:10.1080/01495730151078117

4

T. Zacharia, J. M. Vitek, J. A. Goldak, T. A. DebRoy, M. Rappaz, and H. K. D. H. Bhadeshia, "Modeling of fundamental phenomena in welds," Modelling and Simulation in Materials Science and Engineering, vol. 3, no. 2, pp. 265-288, 1995. doi:10.1088/0965-0393/3/2/009

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £125 +P&P)