Factorial design analytics on effects of material parameter uncertainties in multiphysics modeling of additive manufacturing

Published in Materials
Like

Laser Powder Bed Fusion (L-PBF) is a popular metal additive manufacturing (AM) process that is capable of manufacturing products with complex geometries. Despite its potential, the major challenge for the L-PBF process is the quality inconsistency of products. Efforts have been invested to analyze the root causes of the issue via simulation and modelling, in order to circumvent costly experimentation. However, input parameter uncertainty limits the accuracy and efficacy of the models. Therefore uncertainty quantification (UQ) for multi-physics simulation models has been gaining traction in the recent years. In our study, we focus on UQ for a high-fidelity multi-physics model.

Now, you might be thinking, "Is it worth waiting for 2 or 3 days for one simulation?" It's a trade-off between computational cost and model accuracy, where sacrificing accuracy may save time and resources, but may compromise on the quality of the analysis. Some will choose to reduce computational cost by neglecting certain physical phenomena in their models, and perform UQ on lower-fidelity multi-physics models. Nevertheless, such an approach has limited accuracy and practical application of the insights obtained. Therefore, we have applied UQ on a high-fidelity multi-physics model, which proved to be a powerful source of obtaining useful insights, especially those that cannot be easily discovered in physical experimentation.

Due to the high computational cost of the high-fidelity multi-physics model, the role of Design of Experiments (DOE) becomes vital. We opted for a full factorial approach because we suspected significant interactions between parameters arising from input uncertainties. But what does a statistical interaction like λdγ/dT actually mean in reality?  That was a question that needed an answer, and we attempted to draw a parallel between the statistical interaction and its physical meaning. And guess what?  We found that some interaction effects, like 

and λdγ/dT, were significant and even related to physical phenomena like the Marangoni effect!

At a reasonable computational cost and complexity, our study derived practical and useful insights for the AM community in terms of resource allocation. The combination of multiple statistical approaches, along with physical verification, helped us to work within the constraint of a small data set, and achieve statistical-physical validation for our analysis. Overall, our research article offers valuable insights to the AM community, specifically regarding resource allocation, which has the potential to enhance the L-PBF process. If you're interested in learning more, be sure to check out our latest published article in npj Computational Materials.

Please sign in or register for FREE

If you are a registered user on Research Communities by Springer Nature, please sign in

Subscribe to the Topic

Materials Science
Physical Sciences > Materials Science

Related Collections

With collections, you can get published faster and increase your visibility.

Computational Progress of High Entropy Materials

We invite submissions of papers focusing on the computational progress of high entropy materials.

Publishing Model: Open Access

Deadline: Jul 07, 2024

Transport Mechanisms in Energy Materials

The integration of computational material methods, artificial intelligence technology, and advanced in-situ experimental characterization techniques constitutes a foundational approach for unraveling the microstructural transport mechanisms within energy materials. The recent surge in mechanism elucidation, powered by these integrated methodologies, is widely acknowledged as a pivotal avenue for material innovations, consequently propelling advancements in new energy applications. This collection is dedicated to tracking the latest developments and publishing intriguing investigations pertaining to transport mechanisms within energy materials.

Publishing Model: Open Access

Deadline: Oct 31, 2024