Deep Reinforcement Learning for Multi-Phase Microstructure Design

dc.contributor.authorYang, Jiongzhien
dc.contributor.authorHarish, Srivatsaen
dc.contributor.authorLi, Candyen
dc.contributor.authorZhao, Hengduoen
dc.contributor.authorAntous, Brittneyen
dc.contributor.authorAcar, Pinaren
dc.contributor.departmentMechanical Engineeringen
dc.date.accessioned2021-06-08T15:40:57Zen
dc.date.available2021-06-08T15:40:57Zen
dc.date.issued2021-03-22en
dc.description.abstractThis paper presents a de-novo computational design method driven by deep reinforcement learning to achieve reliable predictions and optimum properties for periodic microstructures. With recent developments in 3-D printing, microstructures can have complex geometries and material phases fabricated to achieve targeted mechanical performance. These material property enhancements are promising in improving the mechanical, thermal, and dynamic performance in multiple engineering systems, ranging from energy harvesting applications to spacecraft components. The study investigates a novel and efficient computational framework that integrates deep reinforcement learning algorithms into finite element-based material simulations to quantitatively model and design 3-D printed periodic microstructures. These algorithms focus on improving the mechanical and thermal performance of engineering components by optimizing a microstructural architecture to meet different design requirements. Additionally, the machine learning solutions demonstrated equivalent results to the physics-based simulations while significantly improving the computational time efficiency. The outcomes of the project show promise to the automation of the design and manufacturing of microstructures to enable their fabrication in large quantities with the utilization of the 3-D printing technology.en
dc.description.notesThe presented work was funded by the NASA Virginia Space Grant Consortium Grant (Project Title: "Deep Reinforcement Learning for De-Novo Computational Design of Meta-Materials").en
dc.description.sponsorshipNASA Virginia Space Grant Consortium Grant (Project Title: "Deep Reinforcement Learning for De-Novo Computational Design of Meta-Materials")en
dc.description.versionPublished versionen
dc.format.mimetypeapplication/pdfen
dc.identifier.doihttps://doi.org/10.32604/cmc.2021.016829en
dc.identifier.eissn1546-2226en
dc.identifier.issn1546-2218en
dc.identifier.issue1en
dc.identifier.urihttp://hdl.handle.net/10919/103697en
dc.identifier.volume68en
dc.language.isoenen
dc.rightsCreative Commons Attribution 4.0 Internationalen
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/en
dc.subjectDeep learningen
dc.subjectreinforcement learningen
dc.subjectmicrostructure designen
dc.titleDeep Reinforcement Learning for Multi-Phase Microstructure Designen
dc.title.serialCMC-Computers Materials & Continuaen
dc.typeArticle - Refereeden
dc.type.dcmitypeTexten
dc.type.dcmitypeStillImageen

Files

Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
TSP_CMC_41868.pdf
Size:
538.46 KB
Format:
Adobe Portable Document Format
Description:
Published version