A machine learning framework to predict local strain distribution and the evolution of plastic anisotropy &amp; fracture in additively manufactured alloys

Muhammad, Waqas; Brahme, Abhijit; Ibragimova, Olga; Kang, Jidong; Inal, Kaan

doi:10.1016/j.ijplas.2020.102867

Cited by 102 publications

(24 citation statements)

References 60 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…For sigmoid, deeper analysis should be applied to interpret the obtained huge error. Indeed, Waqas et al [17] (see Introduction) had obtained similar losses evolution by applying sigmoid and tanh, when they investigated the effect of the choice of the activation function. In their case, a neural network with 8 hidden layers was selected while 2, 3 and 4 hidden layers (using Relu) were here investigated as shown in Fig.…”

Section: R Results and Discussion Esults And Discussionmentioning

confidence: 99%

“…2 and here a is the output value of a given perceptron, f is the activation function (often non-linear), wi the weight of an input connexion i whose value is xi and b the bias. The activation functions are various such as the Rectified Linear Unit (Relu), Logistic (sigmoid), the hyperbolic tangent (tanh), and the softmax (also called softargmax) [17,11]. T Table 1.…”

Section: Ann Model Ann Modelmentioning

confidence: 99%

“…Supervised learning is used in case of known responses and englobes regression algorithms on the one hand and classification ones on the other hand, while unsupervised learning is used in case of unknown response (prediction, recognition, etc.). Regression techniques in AM field have the aim to predict continuous responses (thermal histories [13], melt pool depth [14], deformations [15], etc.). For that, linear or nonlinear regressions, Gaussian Process (GP), Support Vector Machine (SVM), Logistic and Artificial Neural Networks (ANN) are good candidates.…”

Section: Intr Introduction Oductionmentioning

confidence: 99%

“…The model was used to identify appropriate process parameters to get the desirable heat conduction and reduce keyhole mode effects. Waqas et al [17] proposed a model of the deposition of AlSi10Mg alloy by Selective Laser Melting (SLM) in order to predict plastic anisotropy, local strain distribution and failure in tensile tests.…”

Section: Intr Introduction Oductionmentioning

confidence: 99%

See 3 more Smart Citations

Thermal field prediction in DED manufacturing process using Artificial Neural Network

et al. 2021

View full text Add to dashboard Cite

In the last decade, machine learning is increasingly attracting researchers in several scientific areas and, in particular, in the additive manufacturing field. Meanwhile, this technique remains as a black box technique for many researchers. Indeed, it allows obtaining novel insights to overcome the limitation of classical methods, such as the finite element method, and to take into account multi-physical complex phenomena occurring during the manufacturing process. This work presents a comprehensive study for implementing a machine learning technique (artificial neural network) to predict the thermal field evolution during the direct energy deposition of 316L stainless steel and tungsten carbides. The framework consists of a finite element thermal model and a neural network. The influence of the number of hidden layers and the number of nodes in each layer was also investigated. The results showed that an architecture based on 3 or 4 hidden layers and the rectified linear unit as the activation function lead to obtaining a high fidelity prediction with an accuracy exceeding 99%. The impact of the chosen architecture on the model accuracy and CPU usage was also highlighted. The proposed framework can be used to predict the thermal field when simulating multi-layer deposition.

show abstract

Section: R Results and Discussion Esults And Discussionmentioning

confidence: 99%

Section: Ann Model Ann Modelmentioning

confidence: 99%

Section: Intr Introduction Oductionmentioning

confidence: 99%

Section: Intr Introduction Oductionmentioning

confidence: 99%

See 2 more Smart Citations

Thermal field prediction in DED manufacturing process using Artificial Neural Network

et al. 2021

View full text Add to dashboard Cite

show abstract

“…ANN models have also been used to predict material damage under cyclic loading [8]- [11]. Recently, ANN models have also been used to predict local strain distributions [12] and forming limit diagrams (FLD's) [13] for various metallic alloys. Even though ANN models have been used for various applications, literature lacks works in predicting texture dependent stress-strain response.…”

Section: Introductionmentioning

confidence: 99%

Artificial Neural Network Based Crystal Plasticity Framework To Simulate Flow Behavior In Aluminum Alloys

Ali¹

2021

Progress in Canadian Mechanical Engineering. Volume 4

View full text Add to dashboard Cite

Machine learning techniques are used to understand and predict data trends. The flexibility of machine learning approaches such as artificial neural networks (ANN) has seen these approaches being used in various applications. As experimental methods are costly and time-consuming, numerical simulations are often used for their predictive capabilities. In this work, ANN framework is proposed to predict the stress-strain and texture evolution response under simple shear. Stress-strain response from individual crystals were trained and validated with the ANN model to predict the polycrystalline response. Results from the ANN model were compared to experimental simple shear stress-strain results and show good agreement.

show abstract

Effect of Microstructural Evolution on Mechanical Properties and Fracture Modes of AlSi10Mg Blocks Fabricated by Selective Laser Melting after Stress Relief Annealing

Wang

et al. 2022

Adv Eng Mater

View full text Add to dashboard Cite

AlSi10Mg parts fabricated by selective laser melting (SLM) are frequently reported to have heterogeneous microstructures. How this affects the mechanical properties of SLM parts still requires further study. Herein, AlSi10Mg blocks with simple cuboid shape are printed and annealed at 300 °C. Their microstructure changes significantly from edge to center and from top to bottom. Low‐temperature annealing is not sufficient to eliminate structural heterogeneity. Before annealing, the Si‐rich phase is fine and mainly exists in the form of a eutectic network. The as‐printed blocks have high tensile strength, and the fracture surface exhibits a mix of transcellular pits and pores. After annealing, the network structure is completely broken or disappeared in most locations in the blocks. The Si‐rich phase is significantly coarsened and even segregated locally along the melt pool boundaries. Yield strength is reduced by 25%–47%, but elongation values are doubled. Tensile samples are fractured by an along‐the‐pool boundary fracture mode along the boundary. Micropores formed during the printing process are aggregated in the about 200 μm thick outer surface layer of the block. They also have a significant effect on the local elongation both before and after annealing.

show abstract

A machine learning framework to predict local strain distribution and the evolution of plastic anisotropy & fracture in additively manufactured alloys

Cited by 102 publications

References 60 publications

Thermal field prediction in DED manufacturing process using Artificial Neural Network

Thermal field prediction in DED manufacturing process using Artificial Neural Network

Artificial Neural Network Based Crystal Plasticity Framework To Simulate Flow Behavior In Aluminum Alloys

Effect of Microstructural Evolution on Mechanical Properties and Fracture Modes of AlSi10Mg Blocks Fabricated by Selective Laser Melting after Stress Relief Annealing

Contact Info

Product

Resources

About