TY - JOUR
T1 - Machine learning for topology optimization
T2 - Physics-based learning through an independent training strategy
AU - Senhora, Fernando V.
AU - Chi, Heng
AU - Zhang, Yuyu
AU - Mirabella, Lucia
AU - Tang, Tsz Ling Elaine
AU - Paulino, Glaucio H.
N1 - Funding Information:
The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Fernando V. Senhora reports that financial support was provided by Siemens Corporation, Technology.
Funding Information:
The authors acknowledge the financial support from “ Siemens Corporation, Technology, USA ” under the project titled “Deep Learning Enhanced Topology Optimization”, and through a Doctoral Fellowship to FVS. We also acknowledge support from the US National Science Foundation (NSF) under grant # 2105811 .
Publisher Copyright:
© 2022
PY - 2022/8/1
Y1 - 2022/8/1
N2 - The high computational cost of topology optimization has prevented its widespread use as a generative design tool. To reduce this computational cost, we propose an artificial intelligence approach to drastically accelerate topology optimization without sacrificing its accuracy. The resulting AI-driven topology optimization can fully capture the underlying physics of the problem. As a result, the machine learning model, which consists of a convolutional neural network with residual links, is able to generalize what it learned from the training set to solve a wide variety of problems with different geometries, boundary conditions, mesh sizes, volume fractions and filter radius. We train the machine learning model separately from the topology optimization, which allows us to achieve a considerable speedup (up to 30 times faster than traditional topology optimization). Through several design examples, we demonstrate that the proposed AI-driven topology optimization framework is effective, scalable and efficient. The speedup enabled by the framework makes topology optimization a more attractive tool for engineers in search of lighter and stronger structures, with the potential to revolutionize the engineering design process. Although this work focuses on compliance minimization problems, the proposed framework can be generalized to other objective functions, constraints and physics.
AB - The high computational cost of topology optimization has prevented its widespread use as a generative design tool. To reduce this computational cost, we propose an artificial intelligence approach to drastically accelerate topology optimization without sacrificing its accuracy. The resulting AI-driven topology optimization can fully capture the underlying physics of the problem. As a result, the machine learning model, which consists of a convolutional neural network with residual links, is able to generalize what it learned from the training set to solve a wide variety of problems with different geometries, boundary conditions, mesh sizes, volume fractions and filter radius. We train the machine learning model separately from the topology optimization, which allows us to achieve a considerable speedup (up to 30 times faster than traditional topology optimization). Through several design examples, we demonstrate that the proposed AI-driven topology optimization framework is effective, scalable and efficient. The speedup enabled by the framework makes topology optimization a more attractive tool for engineers in search of lighter and stronger structures, with the potential to revolutionize the engineering design process. Although this work focuses on compliance minimization problems, the proposed framework can be generalized to other objective functions, constraints and physics.
KW - 3D convolutional neural network
KW - Large-scale
KW - Machine Learning
KW - Physics-based machine learning
KW - Topology optimization
KW - Training Set
UR - http://www.scopus.com/inward/record.url?scp=85133459264&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85133459264&partnerID=8YFLogxK
U2 - 10.1016/j.cma.2022.115116
DO - 10.1016/j.cma.2022.115116
M3 - Article
AN - SCOPUS:85133459264
SN - 0374-2830
VL - 398
JO - Computer Methods in Applied Mechanics and Engineering
JF - Computer Methods in Applied Mechanics and Engineering
M1 - 115116
ER -