A study on using image based machine learning methods to develop the surrogate models of stamp forming simulations

In the design optimization of metal forming, it is increasingly significant to use surrogate models to analyse the finite element analysis (FEA) simulations. However, traditional surrogate models using scalar based machine learning methods (SBMLMs) fall in short of accuracy and generalizability. This is because SBMLMs fail to harness the location information of the simulations. To overcome these shortcomings, image based machine learning methods (IBMLMs) are leveraged in this paper. The underlying theory of location information, which supports the advantages of IBMLM, is qualitatively interpreted. Based on this theory, a Res-SE-U-Net IBMLM surrogate model is developed and compared with a multi-layer perceptron (MLP) as a referencing SBMLM surrogate model. It is demonstrated that the IBMLM model is advantageous over the MLP SBMLM model in accuracy, generalizability, robustness, and informativeness. This paper presents a promising methodology of leveraging IBMLMs in surrogate models to make maximum use of info from FEA results. Future prospective studies that inspired by this paper are also discussed.

[1]  Axel Schumacher,et al.  Automatic Analysis of Crash Simulations with Dimensionality Reduction Algorithms such as PCA and t-SNE , 2020 .

[2]  Levent Burak Kara,et al.  StressGAN: A Generative Deep Learning Model for 2D Stress Distribution Prediction , 2020, DAC 2020.

[3]  M. Ovsjanikov,et al.  Instant recovery of shape from spectrum via latent space connections , 2020, 2020 International Conference on 3D Vision (3DV).

[4]  Markus Bambach,et al.  A Fast Approach for Optimization of Hot Stamping Based on Machine Learning of Phase Transformation Kinetics , 2020 .

[5]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[6]  Frank Henning,et al.  A machine learning assisted approach for textile formability assessment and design improvement of composite components , 2019, Composites Part A: Applied Science and Manufacturing.

[7]  M. Schneider,et al.  Fast methods for computing centroidal Laguerre tessellations for prescribed volume fractions with applications to microstructure generation of polycrystalline materials , 2020 .

[8]  Changhoon Lee,et al.  Prediction of turbulent heat transfer using convolutional neural networks , 2019, Journal of Fluid Mechanics.

[9]  L. Kärger,et al.  Estimating Optimum Process Parameters in Textile Draping of Variable Part Geometries - A Reinforcement Learning Approach , 2020, Procedia Manufacturing.

[10]  Nils Thuerey,et al.  Deep Learning Methods for Reynolds-Averaged Navier–Stokes Simulations of Airfoil Flows , 2018, AIAA Journal.

[11]  Luise Kärger,et al.  Virtual Product Development Using Simulation Methods and AI , 2019, Lightweight Design worldwide.

[12]  Wei Chen,et al.  Scalable Objective-Driven Batch Sampling in Simulation-Based Design for Models With Heteroscedastic Noise , 2020, Volume 11B: 46th Design Automation Conference (DAC).

[13]  Wei Sun,et al.  A deep learning approach to estimate stress distribution: a fast and accurate surrogate of finite-element analysis , 2018, Journal of The Royal Society Interface.

[14]  Daniel Trippe,et al.  An approach for rapid prediction of textile draping results for variable composite component geometries using deep neural networks , 2019 .

[15]  M. Stein.  Large sample properties of simulations using latin hypercube sampling , 1987 .

[16]  Junfeng Chen,et al.  U-net architectures for fast prediction of incompressible laminar flows , 2019, 1910.13532.

[17]  Hui Yang,et al.  Deep Learning of Variant Geometry in Layerwise Imaging Profiles for Additive Manufacturing Quality Control , 2019, Journal of Manufacturing Science and Engineering.

[18]  Sen Jia,et al.  How Much Position Information Do Convolutional Neural Networks Encode? , 2020, ICLR.

[19]  Jerry Y. H. Fuh,et al.  Thermal field prediction for laser scanning paths in laser aided additive manufacturing by physics-based machine learning , 2020 .

[20]  Ikjin Lee,et al.  Deep Generative Design: Integration of Topology Optimization and Generative Models , 2019, Journal of Mechanical Design.

[21]  Amos Storkey,et al.  Comparing recurrent and convolutional neural networks for predicting wave propagation , 2020, ArXiv.

[22]  Baki Harish,et al.  Topology Optimization Using Convolutional Neural Network , 2020 .

[23]  Enying Li,et al.  Time dependent sheet metal forming optimization by using Gaussian process assisted firefly algorithm , 2018 .

[24]  Tao Zeng,et al.  Reconstruction of Simulation-Based Physical Field by Reconstruction Neural Network Method. , 2018, 1805.00528.

[25]  Wei Li,et al.  Convolutional Neural Networks for Steady Flow Approximation , 2016, KDD.

[26]  K. Bathe.  Finite Element Procedures , 1995 .

[27]  Kaiming He,et al.  Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[28]  Shan Ren,et al.  Metamodel-Based Multi-Objective Reliable Optimization for Front Structure of Electric Vehicle , 2018 .

[29]  J. Beyerer,et al.  Optimisation of manufacturing process parameters using deep neural networks as surrogate models , 2018 .

[30]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[31]  Enrico Zio,et al.  A novel time-dependent system constraint boundary sampling technique for solving time-dependent reliability-based design optimization problems , 2020 .

[32]  Ross B. Girshick,et al.  Mask R-CNN , 2017, 1703.06870.

[33]  Louis J. Durlofsky,et al.  A deep-learning-based surrogate model for data assimilation in dynamic subsurface flow problems , 2019, J. Comput. Phys..

[34]  Trevor A. Dean,et al.  Materials Modelling for Selective Heating and Press Hardening of Boron Steel Panels with Graded Microstructures , 2014 .

[35]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[36]  Yong Peng,et al.  Image-based reconstruction for the impact problems by using DPNNs. , 2019 .

[37]  Jida Huang,et al.  Geometric Deep Learning for Shape Correspondence in Mass Customization by Three-Dimensional Printing , 2020 .

[38]  Geoffrey E. Hinton,et al.  Rectified Linear Units Improve Restricted Boltzmann Machines , 2010, ICML.

[39]  Haotian Feng,et al.  Difference-Based Deep Learning Framework for Stress Predictions in Heterogeneous Media , 2020, ArXiv.

[40]  Constantin Diez.  MACHINE LEARNING PROCESS TO ANALYZE BIG-DATA FROM CRASH SIMULATIONS , 2017 .

[41]  Levent Burak Kara,et al.  TopologyGAN: Topology Optimization Using Generative Adversarial Networks Based on Physical Fields Over the Initial Domain , 2020, Journal of Mechanical Design.

[42]  Levent Burak Kara,et al.  Deep Learning for Stress Field Prediction Using Convolutional Neural Networks , 2018, J. Comput. Inf. Sci. Eng..

[43]  Jiong Tang,et al.  A mode tracking method in modal metamodeling for structures with clustered eigenvalues , 2020 .

[44]  Nils Thuerey,et al.  A Combined Data-driven and Physics-driven Method for Steady Heat Conduction Prediction using Deep Convolutional Neural Networks , 2020, ArXiv.

[45]  Tobias C. Spruegel,et al.  Generic approach to plausibility checks for structural mechanics with deep learning , 2017 .

[46]  Thermal modeling on satellites panels: 3D Deep Learning on different topologies , 2020 .

[47]  Jun Wang,et al.  Data-driven simulation for fast prediction of pull-up process in bottom-up stereo-lithography , 2016, Comput. Aided Des..

[48]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[49]  L. Bian,et al.  Deep Learning-Based Data Fusion Method for In Situ Porosity Detection in Laser-Based Additive Manufacturing , 2020, Journal of Manufacturing Science and Engineering.

[50]  Karthik Duraisamy,et al.  Turbulence Modeling in the Age of Data , 2018, Annual Review of Fluid Mechanics.

[51]  M.-H. Herman Shen,et al.  A Novel Topology Optimization Approach using Conditional Deep Learning , 2019, ArXiv.

[52]  M. P. Brenner,et al.  Perspective on machine learning for advancing fluid mechanics , 2019, Physical Review Fluids.

[53]  Sagar Patel,et al.  3D Topology Optimization using Convolutional Neural Networks , 2018, ArXiv.