Deep learning for topology optimization of 2D metamaterials

Kollmann, Hunter T.; Abueidda, Diab W.; Korić, Seid; Guleryuz, Erman; Sobh, Nahil

doi:10.1016/j.matdes.2020.109098

Cited by 239 publications

(98 citation statements)

References 71 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Numerical modeling in mechanics and material science is not an exception. Various surrogate deep learning data-driven models have been trained to learn and quickly inference the thermal conductivity and advanced manufacturing of composites [14,15], topologically optimized materials and structures [16,17], the fatigue of materials [18], nonlinear material response such as in plasticity and viscoplasticity [19,20], and many other applications.…”

Section: Introductionmentioning

confidence: 99%

Deep Learning Sequence Methods in Multiphysics Modeling of Steel Solidification

Korić

Abueidda

2021

Metals

Self Cite

View full text Add to dashboard Cite

The solidifying steel follows highly nonlinear thermo-mechanical behavior depending on the loading history, temperature, and metallurgical phase fraction calculations (liquid, ferrite, and austenite). Numerical modeling with a computationally challenging multiphysics approach is used on high-performance computing to generate sufficient training and testing data for subsequent deep learning. We have demonstrated how the innovative sequence deep learning methods can learn from multiphysics modeling data of a solidifying slice traveling in a continuous caster and correctly and instantly capture the complex history and temperature-dependent phenomenon in test data samples never seen by the deep learning networks.

show abstract

Section: Introductionmentioning

confidence: 99%

Deep Learning Sequence Methods in Multiphysics Modeling of Steel Solidification

Korić

Abueidda

2021

Metals

Self Cite

View full text Add to dashboard Cite

show abstract

“…Although this method has demonstrated powerful capabilities, it is difficult to exploit the superior properties fully. As a remedy, topology optimization methods have been further developed to design mechanical metamaterials with optimal (or locally optimal) mechanical properties [9][10][11]. Nevertheless, topology optimization may not always be possible obtaining the desired optimal solutions, thereby leading to the imperious demand for new design ideas for the mechanical metamaterial.…”

Section: Introductionmentioning

confidence: 99%

Stacked-origami mechanical metamaterial with tailored multistage stiffness

Wen

Chen

Long

et al. 2021

Preprint

View full text Add to dashboard Cite

Origami-baed metamaterial has shown remarkable mechanical properties rarely found in natural materials, but achieving tailored multistage stiffness is still a challenge. This study proposes a novel zigzag-base stacked-origami (ZBSO) metamaterial with tailored multistage stiffness property based on crease customization and stacking strategies. A high precision finite element (FE) model to identify the stiffness characteristics of the ZBSO metamaterial has been established, and its accuracy is validated by quasi-static compression experiments. Using the verified FE model, we demonstrate that the multistage stiffness of the ZBSO metamaterial can be effectively tailored through two manners, i.e. varying the microstructures (through introducing new creases to the classical Miura origami unit cell) and altering the stacking way. Three strategies are utilized to vary the microstructure, i.e. adding new creases to the right, left, or both sides of the unit cell. We further reveal that the proposed ZBSO metamaterial has several outstanding advantages compared with traditional mechanical metamaterials, e.g. material independent, scale-invariant, lightweight, and excellent energy absorption capacity. The unravelled superior mechanical properties of the ZBSO metamaterials pave the way for the design of the next-generation cellular metamaterials with tailored stiffness properties.

show abstract

“…However, due to prescribed parametrization of the geometry, the design freedom in the presented approaches is limited and hence one does not exploit the full lightweight potential. Kollmann et al (2020) utilized topology optimization to estimate microstructure designs. In order to determine the stiffness matrix, substitute load cases were set up, which were used to optimize different topologies with respect to mass.…”

Section: Introductionmentioning

confidence: 99%

Decomposition and optimization of linear structures using meta models

Krischer

Zimmermann

2021

Struct Multidisc Optim

View full text Add to dashboard Cite

Monolithic optimization of large mechanical systems can be expensive and cumbersome. Drivers of computational cost and integration effort are, e.g., the size of the design problem and the number of different components, models, and disciplines. Distributed optimization schemes decompose large problems into smaller subproblems; however, they typically require intense coordination effort. This paper proposes an approach for complete decoupling by decomposing a monolithic optimization into independent optimization subproblems that can be solved without need for coordination. This is accomplished by sampling the space of component performance, here represented by eigenvalues and eigenvectors of stiffness matrices, and establishing meta models that map the relevant component performance values onto feasibility and mass estimates. The optimization procedure consists of two steps: First, a system optimization problem is solved by assigning stiffness requirements to components that are approximately feasible and mass-optimal. Second, the component optimization problems are solved independently of each other such that stiffness requirements are satisfied. As information on feasibility and mass is provided during system optimization by meta models, the approach will be referred to as informed decomposition. The effectiveness of the approach is demonstrated by minimizing the mass of a simple two-component linear structure subject to a requirement on total stiffness. This is done for three different component models, a beam with constant cross-section, a beam with varying cross-sections, and an arbitrary 2-dimensional body, using parametric and topology optimization, respectively. The approach produces results that are at most 1 % heavier than the results obtained by monolithic optimization.

show abstract

Deep learning for topology optimization of 2D metamaterials

Cited by 239 publications

References 71 publications

Deep Learning Sequence Methods in Multiphysics Modeling of Steel Solidification

Deep Learning Sequence Methods in Multiphysics Modeling of Steel Solidification

Stacked-origami mechanical metamaterial with tailored multistage stiffness

Decomposition and optimization of linear structures using meta models

Contact Info

Product

Resources

About