SUBJECT: Ph.D. Proposal Presentation
   
BY: David Montes De Oca Zapiain
   
TIME: Thursday, April 5, 2018, 9:00 a.m.
   
PLACE: MRDC Building, 4211
   
TITLE: Development of reduced-order models for predicting the plastic deformation of metals employing material knowledge systems.
   
COMMITTEE: Dr. Surya Kalidindi, Chair (ME)
Dr. David McDowell (MSE)
Dr. Hamid Garmestani (MSE)
Dr. Richard W. Neu (ME)
Dr. Hojun Lim (SANDIA National Laboratories)
 

SUMMARY

Metal forming processes plastically deform the material to a design specified shape (e.g. stamping and forming).The severe plastic deformation enhances mechanical properties of metals desirable in structural and technological applications (i.e. yield strength). Computational frameworks that employ numerical approaches provide useful insights about the deformation of the material without the need to perform costly experiments. For this reason, numerical approaches are used to better design and optimize these manufacturing processes. However, in order to obtain accurate and realistic models (or simulations) for these processes it is necessary to account for the effects of material heterogeneities at different length scales on the overall plastic deformation. The strategies to account for these effects used by current computational frameworks have a high computational cost and are extremely time consuming. Therefore, a better strategy is required. A novel, systematic and computationally efficient way to account for these effects is to develop/establish scale bridging tools. Scale bridging tools effectively and efficiently communicate salient information amongst the different scales involved in the form of reduced-order models. Nevertheless, in order for these tools to be valid the reduced-order models need to capture accurately the high-value quantitative connections between the complex material microstructure and its associated plastic deformation. The proposed thesis work will define a robust framework capable of generating the sought-after reduced-order models by leveraging a novel data science approach called Material Knowledge Systems (MKS). The developed framework will yield accurate reduced-order models that realistically predict the plastic response of metals in a computationally efficient manner. As a result, the proposed thesis work will be paving the way forward for the establishment of accurate and computationally efficient multiscale plasticity simulations by defining a robust and practical scale bridging framework for plastic deformations with the aid of the MKS framework.