Session: 04-06-01: AI for Heterogeneous Materials Design, Discovery, and Manufacturing I

Paper Number: 173461

Data-Driven Approach for Use-Inspired Multi-Material Design of Architectured Lattices 

Multimaterial additive manufacturing (MAM) has introduced a new dimension of capability and versatility in the already revolutionary field of 3D printing through enabling fabrication of custom geometries via multiple materials. An important class of fabricated materials are architected strut-based solids, such as octet-truss lattices, truncated octahedron lattices. The versatility of MAM techniques renders an exceedingly high-dimensional set of possible multi-material designs for a given geometry or structure  This vast space of realizable designs, in turn, calls upon a systematic explore-and-select framework that accurately accounts for both the mechanics and the functionality of the resultant product. To this end, a computational multi-material design framework that, for a given geometry, objective, and collection of material choices, distributes an optimal selection of materials is desirable and timely.

We propose a constitutive model-free data-driven-design (D³) approach for optimal designs of multi-material lattices here, where we directly utilize data to find the optimal selection of materials in a use-inspired manner. Our approach relies on reformulating boundary value problems in mechanics as a distance minimization problem in an abstract space of all possible stresses and strains. In this talk, after presenting the general framework for multi-material design, we will focus on the viscoelastic lattices and showcase our approach for designing multi-material architectured lattices with maximal dissipation. In one line of efforts, we demonstrate that D³ enables designing multi-material three-dimensional unit-cells for four different affine deformation patterns. We provided our framework with the dynamic viscoelastic data for 25 different materials as variants of Epoxy, ABS, and TB+. We conduct optimal designs for both single-frequency and multi-frequency cases, and show that a multi-material design, by orchestrating the heterogeneities and corresponding deformations, can mirror the dissipation of a homogeneous lattice made of the most dissipative material. Separately, we investigate multi-material design of a finite lattice under a non-uniform loading with a similar objective of maximizing mechanical dissipation. In contrast with a lattice homogeneously comprised of the most dissipative material, we demonstrate that D³  provides an optimal design with 3-fold enhancement in dissipation. We show that a similar improvement can be achieved by restricting the material designs to three materials only.

We conclude by discussing how our framework naturally generalizes to multi-physics and multi-objective metastructure design, offering a unified, data-driven approach to optimal material selection under complex constraints. This, in turn, is particularly useful for sustainable material design where a given functionality and design is desired while restricting usage of a certain undesirable material and simultaneously maximizing incorporation of a favourite recyclable material without compromising on the functionality.

Presenting Author: Rayehe Karimi Mahabadi Duke University

Presenting Author Biography: Professor (Amir)Hossein Salahshoor is an Assistant Professor in the departments of Civil and Environmental Engineering, and Mechanical Engineering and Material Science at Duke. Before that, he conducted his postdoctoral studies at Caltech. Prior to that he obtained his Ph.D in Aerospace Engineering from Georgia Tech, along with an MS in Mathematics. His research interests broadly lie at the intersection of mechanics of materials and structures, computational and data science, biology, and applied mathematics.

Authors:

Rayehe Karimi Mahabadi Duke University
Hossein Salahshoor N/A