Deep Learning for Stress Field Prediction Using Convolutional Neural Networks

  title={Deep Learning for Stress Field Prediction Using Convolutional Neural Networks},
  author={Zhenguo Nie and Haoliang Jiang and Levent Burak Kara},
This research presents a deep learning based approach to predict stress fields in the solid material elastic deformation using convolutional neural networks (CNN). Two different architectures are proposed to solve the problem. One is Feature Representation embedded Convolutional Neural Network (FR-CNN) with a single input channel, and the other is Squeeze-and-Excitation Residual network modules embedded Fully Convolutional Neural network (SE-Res-FCN) with multiple input channels. Both the tow… 
StressNet - Deep learning to predict stress with fracture propagation in brittle materials
A deep learning model is proposed to predict the entire sequence of maximum internal stress based on fracture propagation and the initial stress data, using the Temporal Independent Convolutional Neural Network (TI-CNN) and the Bidirectional Long Short-term Memory (Bi-LSTM) Network, to reduce computational cost while preserving accuracy.
Deep learning-based surrogate modeling via physics-informed artificial image (PiAI) for strongly coupled multidisciplinary engineering systems
  • Sungkun Hwang, Seung-Kyum Choi
  • Computer Science
    Knowl. Based Syst.
  • 2021
This research proposes a framework of CNN-based deep surrogate model (DSM), developing a novel input structure called physics-informed artificial image (PiAI), which incubates geometry-informed CAD representing physical uncertainties of engineering systems and applies unimodal-based single image inputs to escalate computational efficiency.
Scalability of Learning Tasks on 3D CAE Models Using Point Cloud Autoencoders
A comprehensive evaluation of how to scale geometric deep learning architectures to high-dimensional inputs to allow for an application of state-of-the-art deep learning methods in real-world tasks is provided.
Bayesian neural networks for uncertainty quantification in data-driven materials modeling
Abstract Modern machine learning (ML) techniques, in conjunction with simulation-based methods, present remarkable potential for various scientific and engineering applications. Within the materials…
Accelerating Conceptual Design Analysis of Marine Vehicles through Deep Learning
The unsteady near wake of a self–propelled body in a linearly stratified environment is studied through the novel use of an actuator–line model for the propeller in unsteady, Reynolds– averaged…


Deep learning for determining a near-optimal topological design without any iteration
The performance evaluation results of the integrated network demonstrate that the proposed method can determine a near-optimal structure in terms of pixel values and compliance with negligible computational time.
A deep learning approach to estimate stress distribution: a fast and accurate surrogate of finite-element analysis
This study marks, to the authors' knowledge, the first study that demonstrates the feasibility and great potential of using the DL technique as a fast and accurate surrogate of FEA for stress analysis.
Generic approach to plausibility checks for structural mechanics with deep learning
A method is presented that enables automatic plausibility checks on basis of empirical simulation datasets and is to support development engineers during design accompanying FEA in virtual product development.
A Deep Learning based Approach to Reduced Order Modeling for Turbulent Flow Control using LSTM Neural Networks
Reduced Order Modeling (ROM) for engineering applications has been a major research focus in the past few decades due to the unprecedented physical insight into turbulence offered by high-fidelity…
ImageNet classification with deep convolutional neural networks
A large, deep convolutional neural network was trained to classify the 1.2 million high-resolution images in the ImageNet LSVRC-2010 contest into the 1000 different classes and employed a recently developed regularization method called "dropout" that proved to be very effective.
Neural networks for topology optimization
  • Ivan Sosnovik, I. Oseledets
  • Computer Science, Mathematics
    Russian Journal of Numerical Analysis and Mathematical Modelling
  • 2019
This research introduces convolutional encoder-decoder architecture and the overall approach of solving the layout problem with high performance and demonstrates the ability of the application of the proposed model to other problems.
Deep Residual Learning for Image Recognition
This work presents a residual learning framework to ease the training of networks that are substantially deeper than those used previously, and provides comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth.
Deep learning for topology optimization design
The proposed methodology of using two generative modeling techniques for deep learning is expected to significantly increase the efficiency of topology optimization.
Squeeze-and-Excitation Networks
  • Jie Hu, Li Shen, Gang Sun
  • Computer Science
    2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition
  • 2018
This work proposes a novel architectural unit, which is term the "Squeeze-and-Excitation" (SE) block, that adaptively recalibrates channel-wise feature responses by explicitly modelling interdependencies between channels and finds that SE blocks produce significant performance improvements for existing state-of-the-art deep architectures at minimal additional computational cost.
Computational mechanics enhanced by deep learning
A new method of numericalquadrature for the FEM stiffness matrices is developed by using the proposed method, where a kind of optimized quadrature rule superior in accuracy to the standard Gauss–Legendre quadratures is obtained on the element-by-element basis.