An integrated modeling approach for predicting process maps of residual stress and distortion in a laser weld: A combined CFD-FE methodology

Laser welding has become an important joining methodology within a number of industries for the structural joining of metallic parts. It offers a high power density welding capability which is desirable for deep weld sections, but is equally suited to performing thinner welded joints with sensible amendments to key process variables. However, as with any welding process, the introduction of severe thermal gradients at the weld line will inevitably lead to process-induced residual stress formation and distortions. Finite element (FE) predictions for weld simulation have been made within academia and industrial research for a number of years, although given the fluid nature of the molten weld pool, FE methodologies have limited capabilities. An improvement upon this established method would be to incorporate a computational fluid dynamics (CFD) model formulation prior to the FE model, to predict the weld pool shape and fluid flow, such that details can be fed into FE from CFD as a starting condition. The key outputs of residual stress and distortions predicted by the FE model can then be monitored against the process variables input to the model. Further, a link between the thermal results and the microstructural properties is of interest. Therefore, an empirical relationship between lamellar spacing and the cooling rate was developed and used to make predictions about the lamellar spacing for welds of different process parameters. Processing parameter combinations that lead to regions of high residual stress formation and high distortion have been determined, and the impact of processing parameters upon the predicted lamellar spacing has been presented