Variation Simulation of Welded Assemblies Using a Thermo-Elastic Finite Element Model

S. Lorin, C. Cromvik, F. Edelvik, L. Lindkvist, R. Söderberg, Journal of Computing and Information Science in Engineering, Apr 28, 2014, 14(3), 031003, 6 pages.


Every series of manufactured products has geometric variation. Variation can lead to products that are difficult to assemble or products not fulfilling functional or aesthetic requirements. In this paper, we will consider the effects of welding in variation simulation. Earlier work that has been combining variation simulation with welding simulation has either applied distortion based on nominal welding conditions onto the variation simulation result, hence loosing combination effects, or has used transient thermo-elasto-plastic simulation, which can be very time consuming since the number of runs required for statistical accuracy can be high. Here, we will present a new method to include the effects of welding in variation simulation. It is based on a technique that uses a thermo-elastic model, which previously has been shown to give distortion prediction within reasonable accuracy. This technique is suited for variation simulations due to the relative short computation times compared to conventional transient thermo-elasto-plastic simulations of welding phenomena. In a case study, it is shown that the presented method is able to give good predictions of both welding distortion and variation of welding distortions compared to transient thermo-elasto-plastic simulations.


This work is part of the ProViking project “Variation Simulation for Light Weight Assemblies,” financed by Swedish Foundation for Strategic Research carried out as a part of the Wingquist Laboratory VINN Excellence Centre research program and Chalmers Production Area of Advance. The support is greatly acknowledged.

Authors and Affiliations

  • S. Lorin, Department of Product and Production Development, Chalmers University of Technology
  • C. Cromvik, Fraunhofer-Chalmers Centre for Industrial Mathematics
  • F. Edelvik, Fraunhofer-Chalmers Centre for Industrial Mathematics
  • L. Lindkvist, Department of Product and Production Development, Chalmers University of Technology
  • R. Söderberg, Department of Product and Production Development, Chalmers University of Technology

Photo credits: Nic McPhee