Details
| Originalsprache | Englisch |
|---|---|
| Aufsatznummer | 2142013 |
| Seitenumfang | 15 |
| Fachzeitschrift | International Journal of Computational Methods |
| Jahrgang | 19 |
| Ausgabenummer | 8 |
| Frühes Online-Datum | 13 Juni 2022 |
| Publikationsstatus | Veröffentlicht - 1 Okt. 2022 |
Abstract
This paper describes a neural network (NN)-based topology optimization approach for designing microstructures. The design variables are the NN weights and biases used to describe the density field, which is independent of element meshes. The number of design variables and gray elements is reduced substantially, and no filtering is necessary. Three numerical examples are provided to demonstrate the efficacy of the proposed method, namely, maximum shear modulus, maximum bulk modulus, and negative Poisson's ratio.
ASJC Scopus Sachgebiete
- Informatik (insg.)
- Informatik (sonstige)
- Mathematik (insg.)
- Computational Mathematics
Zitieren
- Standard
- Harvard
- Apa
- Vancouver
- BibTex
- RIS
in: International Journal of Computational Methods, Jahrgang 19, Nr. 8, 2142013, 01.10.2022.
Publikation: Beitrag in Fachzeitschrift › Artikel › Forschung › Peer-Review
}
TY - JOUR
T1 - Material Design with Topology Optimization Based on the Neural Network
AU - Li, Bin
AU - Guo, Hongwei
AU - Zhuang, Xiaoying
PY - 2022/10/1
Y1 - 2022/10/1
N2 - This paper describes a neural network (NN)-based topology optimization approach for designing microstructures. The design variables are the NN weights and biases used to describe the density field, which is independent of element meshes. The number of design variables and gray elements is reduced substantially, and no filtering is necessary. Three numerical examples are provided to demonstrate the efficacy of the proposed method, namely, maximum shear modulus, maximum bulk modulus, and negative Poisson's ratio.
AB - This paper describes a neural network (NN)-based topology optimization approach for designing microstructures. The design variables are the NN weights and biases used to describe the density field, which is independent of element meshes. The number of design variables and gray elements is reduced substantially, and no filtering is necessary. Three numerical examples are provided to demonstrate the efficacy of the proposed method, namely, maximum shear modulus, maximum bulk modulus, and negative Poisson's ratio.
KW - homogenization
KW - material design
KW - neural network
KW - Topology optimization
UR - http://www.scopus.com/inward/record.url?scp=85132694515&partnerID=8YFLogxK
U2 - 10.1142/S0219876221420135
DO - 10.1142/S0219876221420135
M3 - Article
AN - SCOPUS:85132694515
VL - 19
JO - International Journal of Computational Methods
JF - International Journal of Computational Methods
SN - 0219-8762
IS - 8
M1 - 2142013
ER -