COMPUTATIONAL WELDING MECHANICS

Implications of real-time CWM

If our prediction that CWM can be done in real-time is correct then CWM is likely to be used routinely in industry. This raises the question of what will be the impact of routine use of CWM in industry. We believe this will dramatically change both engineering and research in the three main components of welding technology for all types of welded structures; materials engineering, structural engineering and weld process development, [7].

Materials engineering will be able to simulate microstructure evolution much more accurately. While research on the evolution of microstructures is well advanced, research on the difficult issue of predicting or estimating material properties for a given microstructure with particular emphasis on failure mechanisms is just beginning. The other hard materials research issues will involve local bifurcations such as nucleation of phases, shear bands and porosity; (what materials engineers call nucleation, mathematicians often call bifurcations). These are the three fundamental bifurcations in material engineering. Shear bands involve only deviatoric stress and strain. Porosity involves only volumetric stress and strain. The computational mechanics of these bifurcations is a fairly hard research issue that has been almost totally ignored in the welding literature. It will become a major research topic in materials engineering for welding.

Structural engineering for welded structures will focus on the life cycle of welded structures from conceptual design, to manufacture, in-service use, maintenance and decommissioning. While life cycle engineering of welded structures is not novel, CWM is usually ignored. For example, while the role of residual stresses in welded structures has long been recognized, it has seldom been included in structural analysis. We believe that CWM will enable the effects of residual stress and weld microstructure to be integrated into the life­cycle engineering process. More research will focus on buckling of welded structures. The hardest CWM research issues will be local failure modes due to shear band or porosity formation, i. e., ductile fracture. We expect CWM to become an integral part of structural engineering of welded structures.

The third component of welding technology will be weld process development. The focus will be on the weld pool, the arc and laser physics. This involves the hardest research issue of all - turbulence. Because turbulence is a chaotic phenomenon, CWM of weld processes is likely to strive to resolve essentially small process variations and short time behavior.

Perhaps the most important change arising from the routine use of CWM in industry is that CWM will become closely tied to experiments, including experiments on real production structures. In the past, it has been too expensive to do experiments to validate CWM. If CWM is used by industry, in a sense the experiments become free. This will lead to tight coupling between experimental data and CWM analysis. This in turn will allow both experiments and CWM to be highly optimized and validated. When this stage is reached, CWM will have become a mature technology and welding technology will have a much stronger science base.