Taming Connectedness in Machine-Learning-Based Topology Optimization with Connectivity Graphs

Article Engineering, Multidisciplinary

A deep convolutional neural network for topology optimization with perceptible generalization ability

Dalei Wang et al.

Summary: This article introduces a deep convolutional neural network for structural topology optimization with perceptible generalization ability. It significantly reduces computation cost while maintaining accuracy in design solutions. Furthermore, the model can achieve solutions with a certain accuracy even when boundary conditions are not included in the training dataset.

ENGINEERING OPTIMIZATION (2022)

添加到收藏夹

Review Computer Science, Interdisciplinary Applications

On the use of artificial neural networks in topology optimisation

Rebekka V. Woldseth et al.

Summary: The question of how artificial intelligence methods can improve traditional frameworks for topology optimization has gained attention in the past few years. While different model variations have been proposed with varying levels of success, few significant breakthroughs have been achieved. The literature tends to have a strong belief in the magical capabilities of artificial intelligence, leading to misunderstandings about its limitations. This article presents a critical review of the current state of research in this field and provides recommendations for future scientific progress.

STRUCTURAL AND MULTIDISCIPLINARY OPTIMIZATION (2022)

添加到收藏夹

Article Engineering, Mechanical

TopologyGAN: Topology Optimization Using Generative Adversarial Networks Based on Physical Fields Over the Initial Domain

Zhenguo Nie et al.

Summary: TopologyGAN, a new data-driven topology optimization model, utilizes various physical fields as inputs to achieve better accuracy on test problems compared to the baseline cGAN.

JOURNAL OF MECHANICAL DESIGN (2021)

添加到收藏夹

Article Computer Science, Software Engineering

Real-Time Topology Optimization in 3D via Deep Transfer Learning

Mohammad Mahdi Behzadi et al.

Summary: The literature on topology optimization has expanded rapidly in the past two decades, presenting various methods using shape and topological derivatives or evolutionary algorithms. The main challenge lies in the high computational cost associated with 3D topology optimization problems. Researchers have introduced a transfer learning method based on convolutional neural networks to improve efficiency and accuracy for real-time 3D design exploration.

COMPUTER-AIDED DESIGN (2021)

添加到收藏夹

Article Computer Science, Software Engineering