Machine Learning\u2013Based Reduce Order Crystal Plasticity Modeling for ICME Applications

Mengfei Yuan,Bryce Meredig,Sean Paradiso,Stephen R Niezgoda

doi:10.1007/s40192-018-0123-x

Abstract

Crystal plasticity simulation is a widely used technique for studying the deformation processing of polycrystalline materials. However, inclusion of crystal plasticity simulation into design paradigms such as integrated computational materials engineering (ICME) is hindered by the computational cost of large-scale simulations. In this work, we present a machine learning (ML) framework using the material information platform, Open Citrination, to develop and calibrate a reduced order crystal plasticity model for face-centered cubic (FCC) polycrystalline materials, which can be both rapidly exercised and easily inverted. The reduced order model takes crystallographic texture, constitutive model parameters, and loading condition as inputs and returns the stress-strain curve and final texture. The model can also be inverted and take a stress-strain curve, loading condition, and final texture as inputs and return the initial texture and constitutive model parameters as outputs. Principal component analysis (PCA) is used to develop an efficient description of the crystallographic texture. A viscoplastic self-consistent (VPSC) crystal plasticity solver is used to create the training data by modeling the stress-strain behavior and evolution of texture during deformation processing.

Highlights

In recent decades, the development of computationally aided design methodologies has revolutionized the way productsThe Ohio State University, Columbus, OH 43210, USA are manufactured, from first prototype to final processing, assembly, and testing
As mentioned in “Reduced Order Representation of Crystallographic Texture”, both initial and final orientation distribution function (ODF) are represented by the first 16 principal components (PCs) and the stress-strain behaviors are represented by ten selected points on stress-strain curves
The middle pole figure shows the viscoplastic self-consistent (VPSC) predicted texture resulting from the τ -fiber ODF under a shear loading condition, while the right one shows the predicted ODF from the reduced order model

Summary

Introduction

The development of computationally aided design methodologies has revolutionized the way productsThe Ohio State University, Columbus, OH 43210, USA are manufactured, from first prototype to final processing, assembly, and testing. The frontier is the extension of computational design to include material structure, properties, and processing as critical design variables, which can be optimized to deliver superior material and component-scale performance, rather than constraints on the design process. This is the fundamental goal of Integrated Computational Materials Engineering (ICME) [1]. While offering unprecedented fidelity and predictive power, spatially and temporally resolved material simulation tools can be too computationally expensive or require too much calibration data to be effectively used within the ICME framework [2, 3]. Given that the statistical confidence in the final design

Methods

Results

Conclusion