Deep learning prediction of stress fields in additively manufactured metals with intricate defect networks

Croom, Brendan P.; Berkson, Michael A.; Mueller, Robert; Presley, Michael; Storck, Steven

doi:10.1016/j.mechmat.2021.104191

Cited by 30 publications

(14 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently artificial neural network (ANN) has been utilized as an effective tool to predict an objective variable from multiple factors influencing the variable. Predictions of mechanical properties using ANN models have been reported in metals and alloys, [6][7][8][9][10][11][12][13] ceramics, 14 and metal-ceramic composites. [15][16][17] In these studies, mechanical properties as objective variables were regression-predicted from features such as material compositions, processing parameters, and so forth.…”

Section: Introductionmentioning

confidence: 99%

Fracture toughness evaluation of silicon nitride from microstructures via convolutional neural network

et al. 2022

View full text Add to dashboard Cite

The fracture toughness of silicon nitride (Si3N4) ceramics was evaluated directly from their microstructures via deep learning using convolutional neural network models. Totally 156 data sets containing microstructural images and relative densities were prepared from 45 types of Si3N4 samples as input feature quantities (IFQs) and were correlated to the fracture toughness as an objective variable. The data sets were divided into two groups. One was used for training, resulting in the creation of regression models for two kinds of IFQs: the microstructures only and a combination of the microstructures and the relative densities. The other group was used for testing the validity of the created models. As a result, the determination coefficient was approximately 0.8 even when using only the microstructures as the IFQs and was further improved when adding the relative densities. It was revealed that the fracture toughness of Si3N4 ceramics was well evaluated from their microstructures.

show abstract

Section: Introductionmentioning

confidence: 99%

Fracture toughness evaluation of silicon nitride from microstructures via convolutional neural network

et al. 2022

View full text Add to dashboard Cite

show abstract

“…S3 and S4 . While often equivalent stress distributions, such as von Mises stress, are adopted as single output 45 , 48 , our model learns not only the whole stress tensor (i.e., multiple components) field but also the corresponding deformed shape (Fig. 2 C).…”

Section: Resultsmentioning

confidence: 99%

“…Although stress and strain fields in material systems with variable base material composition exhibiting complex behavior could reasonably be predicted by pixel-based ML models using high resolution images 45 , 47 , 48 , accurately predicting also complex deformed shapes would be computationally costly. To face this challenge using our GNN model, as an example we focus here on the formation of wrinkled interfaces (i.e., instability) in soft layered composites 56 .…”

Section: Resultsmentioning

confidence: 99%

“…We show that GNNs can learn loading conditions correlating deformed shape and stress field as well as physical relationships between material’s structure, and stress (strain) and deformation field. While image-based ML models, such as convolutional neural networks, variational autoencoders, and generative adversarial networks have been widely used to predict physical fields in hierarchical composites 53 , perforated structures 54 , additively manufactured microstructures with defects 45 , and heterogenous microstructures 42 , 48 , the current work presents a more flexible and general ML framework for the prediction of deformed shapes, stress and strain fields with GNNs.

Figure 1 Schematics of the proposed ML model.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Predicting stress, strain and deformation fields in materials and structures with graph neural networks

Maurizi

Gao

Berto

2022

Sci Rep

View full text Add to dashboard Cite

Developing accurate yet fast computational tools to simulate complex physical phenomena is a long-standing problem. Recent advances in machine learning have revolutionized the way simulations are approached, shifting from a purely physics- to AI-based paradigm. Although impressive achievements have been reached, efficiently predicting complex physical phenomena in materials and structures remains a challenge. Here, we present an AI-based general framework, implemented through graph neural networks, able to learn complex mechanical behavior of materials from a few hundreds data. Harnessing the natural mesh-to-graph mapping, our deep learning model predicts deformation, stress, and strain fields in various material systems, like fiber and stratified composites, and lattice metamaterials. The model can capture complex nonlinear phenomena, from plasticity to buckling instability, seemingly learning physical relationships between the predicted physical fields. Owing to its flexibility, this graph-based framework aims at connecting materials’ microstructure, base materials’ properties, and boundary conditions to a physical response, opening new avenues towards graph-AI-based surrogate modeling.

show abstract

“…The use of machine learning has already shown its efficiency in conjunction with additive technologies 43 , 51 , 52 . Currently, there are some studies aimed at applying machine learning in the field of mechanical testing of plastic and metal samples obtained using additive technologies 53 – 59 , including the manufacture of reactor parts 60 . Possible applications can be separated into three groups: a design for additive manufacturing, the manufacturing process itself, and search for structure–property/printing parameter-property relationships 43 .…”

Section: Introductionmentioning

confidence: 99%

Integration of thermal imaging and neural networks for mechanical strength analysis and fracture prediction in 3D-printed plastic parts

Boiko

Korabelnikova

Gordeev

et al. 2022

Sci Rep

View full text Add to dashboard Cite

Additive manufacturing demonstrates tremendous progress and is expected to play an important role in the creation of construction materials and final products. Contactless (remote) mechanical testing of the materials and 3D printed parts is a critical limitation since the amount of collected data and corresponding structure/strength correlations need to be acquired. In this work, an efficient approach for coupling mechanical tests with thermographic analysis is described. Experiments were performed to find relationships between mechanical and thermographic data. Mechanical tests of 3D-printed samples were carried out on a universal testing machine, and the fixation of thermal changes during testing was performed with a thermal imaging camera. As a proof of concept for the use of machine learning as a method for data analysis, a neural network for fracture prediction was constructed. Analysis of the measured data led to the development of thermographic markers to enhance the thermal properties of the materials. A combination of artificial intelligence with contactless nondestructive thermal analysis opens new opportunities for the remote supervision of materials and constructions.

show abstract

Deep learning prediction of stress fields in additively manufactured metals with intricate defect networks

Cited by 30 publications

References 45 publications

Fracture toughness evaluation of silicon nitride from microstructures via convolutional neural network

Fracture toughness evaluation of silicon nitride from microstructures via convolutional neural network

Predicting stress, strain and deformation fields in materials and structures with graph neural networks

Integration of thermal imaging and neural networks for mechanical strength analysis and fracture prediction in 3D-printed plastic parts

Contact Info

Product

Resources

About