Bayesian texture optimization using deep neural network-based numerical material test

Kamijyo, Ryunosuke; Ishii, Akimitsu; Coppieters, Sam; Yamanaka, Akinori

doi:10.1016/j.ijmecsci.2022.107285

Cited by 13 publications

(1 citation statement)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The surface texture provides a cost-effective alternative for improving interfacial properties as it obviates the necessity for component redesign and can be easily implemented in commercial and advanced technologies. − Modern micromachining technology has made it possible to precisely control the shape and size of surface textures. − Thus, the research of surface textures has received a significant boost in interfacial science over the past decade . In turn, the discovery of the mechanisms of surface textures promotes their application development .…”

Section: Introductionmentioning

confidence: 99%

Bionic Structure Inspired by Tree Frogs to Enhance Damping Performance

Yang

Wiercigroch

et al. 2023

ACS Appl. Mater. Interfaces

View full text Add to dashboard Cite

Magnetic fluid shock absorbers (MFSAs) have been successfully utilized to eliminate microvibrations of flexible spacecraft structures. The method of enhancing the damping efficiency of MFSAs has always been a critical issue. To address this, we drew inspiration from the tree frog’s toe pads, which exhibit strong friction due to their unique surface structure. Using 3D printing, we integrated bionic textures copied from tree frog’s toe pad surfaces onto MFSAs, which is the first time to combine bionic design and MFSAs. Additionally, this is also the first time that surface textures have been applied to MFSAs. However, we also had to consider practical engineering applications and manufacturing convenience, so we modified the shape of bionic textures. To do so, we used an edge extraction algorithm for image processing and obtained recognition results. After thorough consideration, we chose hexagon as the shape of surface textures instead of bionic textures. For theoretical analysis, a magnetic field–flow field coupling dynamic model for MFSAs was built for the first time to simulate the magnetic fluid (MF) flow in one oscillation cycle. Using this model, the flow rate contours of the MF were obtained. It was observed that textures cause vortexes to form in the MF layer, which produced an additional velocity field. This increased the shear rate, ultimately leading to an increase in flow resistance. Finally, we conducted vibration reduction experiments and estimated damping characteristics of the proposed MFSAs to prove the effectiveness of both bionic texture and hexagon surface textures. Fortunately, we concluded that hexagon surface textures not only improve the damping efficiency of MFSAs but also require less MF mass.

show abstract

Section: Introductionmentioning

confidence: 99%

Bionic Structure Inspired by Tree Frogs to Enhance Damping Performance

Yang

Wiercigroch

et al. 2023

ACS Appl. Mater. Interfaces

View full text Add to dashboard Cite

show abstract

A multi-task learning-based optimization approach for finding diverse sets of microstructures with desired properties

et al. 2023

View full text Add to dashboard Cite

Optimization along the chain processing-structure-properties-performance is one of the core objectives in data-driven materials science. In this sense, processes are supposed to manufacture workpieces with targeted material microstructures. These microstructures are defined by the material properties of interest and identifying them is a question of materials design. In the present paper, we addresse this issue and introduce a generic multi-task learning-based optimization approach. The approach enables the identification of sets of highly diverse microstructures for given desired properties and corresponding tolerances. Basically, the approach consists of an optimization algorithm that interacts with a machine learning model that combines multi-task learning with siamese neural networks. The resulting model (1) relates microstructures and properties, (2) estimates the likelihood of a microstructure of being producible, and (3) performs a distance preserving microstructure feature extraction in order to generate a lower dimensional latent feature space to enable efficient optimization. The proposed approach is applied on a crystallographic texture optimization problem for rolled steel sheets given desired properties.

show abstract

Machine learning-based fatigue life prediction of metal materials: Perspectives of physics-informed and data-driven hybrid methods

Wang

Gong

et al. 2023

Engineering Fracture Mechanics

View full text Add to dashboard Cite

Bayesian texture optimization using deep neural network-based numerical material test

Cited by 13 publications

References 39 publications

Bionic Structure Inspired by Tree Frogs to Enhance Damping Performance

Bionic Structure Inspired by Tree Frogs to Enhance Damping Performance

A multi-task learning-based optimization approach for finding diverse sets of microstructures with desired properties

Machine learning-based fatigue life prediction of metal materials: Perspectives of physics-informed and data-driven hybrid methods

Contact Info

Product

Resources

About