We investigate the formation of stress hotspots in polycrystalline materials under uniaxial tensile deformation by integrating full field crystal plasticity based deformation models and machine learning techniques to gain data driven insights about microstructural properties. Synthetic 3D microstructures are created representing single phase equiaxed microstructures for generic copper alloys. Uniaxial tensile deformation is simulated using a 3-D full-field, image-based Fast Fourier Transform (FFT) technique with rate-sensitive crystal plasticity, to get local micromechanical fields (stress and strain rates). Stress hotspots are defined as the grains having stress values above the 90th percentile of the stress distribution. Hotspot neighborhoods are then characterized using metrics that reflect local crystallography, geometry, and connectivity. This data is used to create input feature vectors to train a random forest learning algorithm, which predicts the grains that will become stress hotspots. We are able to achieve an area under the receiving operating characteristic curve (ROC-AUC) of 0.74 for face centered cubic materials modeled on generic copper alloys. The results show the power and the limitations of the machine learning approach applied to the polycrystalline grain networks.
This paper describes our approach to the Bosch production line performance challenge run by Kaggle.com. Maximizing the production yield is at the heart of the manufacturing industry. At the Bosch assembly line, data is recorded for products as they progress through each stage. Data science methods are applied to this huge data repository consisting records of tests and measurements made for each component along the assembly line to predict internal failures. We found that it is possible to train a model that predicts which parts are most likely to fail. Thus a smarter failure detection system can be built and the parts tagged likely to fail can be salvaged to decrease operating costs and increase the profit margins.
The first step in constructing a machine learning model is defining the features of the data set that can be used for optimal learning. In this work we discuss feature selection methods, which can be used to build better models, as well as achieve model interpretability. We applied these methods in the context of stress hotspot classification problem, to determine what microstructural characteristics can cause stress to build up in certain grains during uniaxial tensile deformation. The results show how some feature selection techniques are biased and demonstrate a preferred technique to get feature rankings for physical interpretations.
Stress hotspots are regions of stress concentrations that form under deformation in polycrystalline materials. We use a machine learning approach to study the effect of preferred slip systems and microstructural features that reflect local crystallography, geometry, and connectivity on stress hotspot formation in hexagonal close packed materials under uniaxial tensile stress. We consider two cases: one without any preferred slip systems with a critically resolved shear stress (CRSS) ratio of 1:1:1, and a second with CRSS ratio 0.1:1:3 for basal: prismatic: pyramidal slip systems. Random forest based machine learning models predict hotspot formation with an AUC (area under curve) score of 0.82 for the Equal CRSS and 0.81 for the Unequal CRSS cases. The results show how data driven techniques can be utilized to predict hotspots as well as pinpoint the microstructural features causing stress hotspot formation in polycrystalline microstructures.
Variant selection during displacive transformation is popularly monitored by assessing the pole figure or orientation density function of the bulk texture or pole figures from single prior austenite grains. It can also be done by measuring the physical orientation of α′ (bainite or marteniste) plates in the microstructure. It is known that when variants of α′ form randomly, i.e. there is no variant selection, the shear strain associated with each α′ plate gets canceled and transformation strain becomes equal to the volume strain. Under this condition, orientation of sheaves of α′ also shows random distribution. In this work, it has been shown how transformation strain and orientation of sheaves change when variant selection occurs. Mathematical models have already been developed to calculate the transformation strain under various loading conditions; extensive experiments have been conducted during the course of this work to show how the nature of externally applied stress influences the transformation strain. In addition to this, it has also been shown that the variant selection might leave its trace in the microstructure by affecting the alignment of the sheaves under externally applied stress and strain. Mathematical models have been developed to show that variant selection has negligible influence on physical orientation of α′ plates when transformation occurs under external stress, but prior plastic deformation can dictate the orientation to a great extent.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.