Materials Knowledge Systems for Computationally Efficient Bi-directional Scale-bridging in Polycrystal Plasticity
Abstract
Materials Knowledge Systems for Computationally Efficient Bi-directional Scale-bridging in Polycrystal Plasticity Most materials being explored for structural applications in advanced technologies continue to be metals (e.g., Ti alloys in jet engines and bone implants, Advanced High Strength Steels and Mg alloys in lightweight automobiles, Al alloys in aerospace frames, Zr alloys in nuclear industry), whose internal structures at the mesoscale are polycrystalline. Most currently used predictive tools in simulating wrought alloy manufacturing processes are typically based on finite element models that employ phenomenological constitutive theories for the material response. Such models do not account for important details of the mesoscale material structure (e.g., local anisotropy of the material, spatial distribution of defect densities), and lack the fidelity needed to successfully design the manufacturing path for optimal performance of the final product. On the other hand, physics-based crystal plasticity models have enjoyed remarkable success in predicting anisotropic mechanical response of several polycrystalline metals and in predicting the concurrent evolution of certain details of the mesoscale material structure in finite plastic deformation. However, crystal plasticity based (finite element) simulation tools are extremely computationally expensive, and have not yet been adopted broadly by the metal working industry. This proposal aims to develop a new generation of highly efficient spectral crystal plasticity based finite element tools for simulating and designing deformation processing operations in polycrystalline metals. A salient feature of these new tools will be that they will capture heterogeneity of deformation fields at two different length scales – the macroscale and the mesoscale (at the level of individual crystals or grains in a polycrystal). The governing field equations at the macroscale will be solved using the well established finite element models, while the computations at the mesoscale will be handled using a novel data science approach called Materials Knowledge Systems (MKS), developed recently by the PI’s research group. The MKS approach has enjoyed tremendous success in building highly accurate and computationally efficient meta-models for localization (i.e., mesoscale spatial distribution of a macroscale imposed field such as stress or strain rate) in simulating a number of different multi-scale materials phenomena. MKS derives its accuracy from the fact that it is calibrated to results from previously established numerical models for the phenomena of interest, while its computational efficiency comes from the use of Fast Fourier transforms and spectral representations. The primary goal of the proposed research is to develop new MKS based finite element crystal plasticity simulation tools capable of tracking the important details of the evolution of the local mesoscale structure at any material point in a polycrystalline sample subjected to large heterogeneous plastic deformation at the macroscale, while using only modest computational resources. A second equally important goal is to critically validate the predictive capabilities of these new tools with measurements in a rigorous statistical framework. The proposed activity represents a cross-disciplinary effort involving materials science, continuum mechanics, applied mathematics, statistics, data science, and computational science. The central ideas explored in this research are broadly applicable to developing computationally efficient multi-scale simulation tools for a number of materials phenomena beyond the ones described in this proposal. Integration of computation and data sciences into the current practices employed by materials specialists is expected to completely transform how new advanced materials will be designed and deployed in the future.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Aug 12, 2016
- Source ID
- N000141512478
Entities
People
- Surya R. Kalidindi
Organizations
- Georgia Tech Research Corporation
- Office of Naval Research
- United States Navy