This presentation was made at CAASE18, The Conference on Advancing Analysis & Simulation in Engineering. CAASE18 brought together the leading visionaries, developers, and practitioners of CAE-related technologies in an open forum, to share experiences, discuss relevant trends, discover common themes, and explore future issues.

Resource Abstract

The presentation will highlight how new technologies, including the strain-free element activation that follows the physical, space- and time-dependent material deposition (as occurs in additive manufacturing), known as "element progressive activation" also benefits simulation of traditional welding processes, making it much more efficient. “Element progressive activation” (EPA) improves solver preprocessing and analysis performance significantly, and is already used with the Abaqus solver on the 3DExperience platform for additive manufacturing simulations. With EPA, strain-free element activation is no longer an analysis step-dependent feature. It is therefore superior to “model change” (in Abaqus solver) or equivalent approach traditionally used in welding simulations. EPA and the corresponding new model input structures also make for an efficient implementation of a continuous moving torch associated with many welding processes to capture the flux-based energy input, whether it is through well-known Goldak double ellipsoid model or other moving heat source models commonly used in welding simulations. With some automation of pre-processing, it is relatively straight forward to setup a complex 3-D welding analysis model that takes advantage of these new technologies, making realistic simulation of welding processes possible. The presentation will also include a discussion of methods that are more approximate than the moving torch methods, but are more practical and efficient, especially for large models, while still providing acceptable results depending on the simulation objective (e.g. prediction of distortion vs. accurate residual stresses). The 3D “chunking” option, where the bead deposition is approximated as a series of discrete "chunks" (the number of chunks typically controlling the solution accuracy) while still using flux-based energy input to match the torch power, will be illustrated as one such efficient alternative.

It is well known in the welding community, as evidenced from the large number of articles in the published literature, that advanced simulation capabilities like those that will be discussed in the presentation provide an efficient and cost-effective alternative to expensive experimental trial-and-error approach to understanding and optimizing the welding processes. The presentation will conclude with sample simulation results from several example problems in welding as well as validation of models by comparing with experimental results using select, published benchmark problems. A comparative study of the performance of the different modeling options (continuous vs. "chunking") for representative welding problems will also be discussed.