scholarly journals Atomic-position independent descriptor for machine learning of material properties

2018 ◽  
Vol 98 (21) ◽  
Author(s):  
Ankit Jain ◽  
Thomas Bligaard
Keyword(s):  
Machine Learning ◽  
Material Properties ◽  
Atomic Position

Using Artificial Intelligence to Estimate Nonlinear Resilient Modulus Parameters from Common Index Properties

2021 ◽  
pp. 036119812110237
Author(s):  
Laura Camarena
Keyword(s):  
Machine Learning ◽  
Material Properties ◽  
Resilient Modulus ◽  
Nonlinear Behavior ◽  
Machine Learning Techniques ◽  
Design Parameters ◽  
Nonlinear Parameters ◽  
State Highway ◽  
Learning Techniques ◽  
State Highway Agencies

The Mechanistic–Empirical Pavement Design Guide (MEPDG) considers a hierarchical approach to determine the input values necessary for most design parameters. Level 1 requires site-specific measurement of the material properties from laboratory testing, whereas other levels make use of equations developed from regression models to estimate the material properties. Resilient modulus is a mechanical property that characterizes the unbound and subgrade materials under loading that is essential for the mechanistic design of pavements. The MEPDG resilient modulus model makes use of a three-parameter constitutive model to characterize the nonlinear behavior of the geomaterials. As the resilient modulus tests are complex, expensive, and require lengthy preparation time, most state highway agencies are unlikely to implement them as routine daily applications. Therefore, it is imperative to make use of models to calculate these nonlinear parameters. Existing models to determine these parameters are frequently based on linear regression. With the development of machine learning techniques, it is feasible to develop simpler equations that can be used to estimate the nonlinear parameters more accurately. This study makes use of the Long-Term Pavement Performance database and machine learning techniques to improve the equations utilized to determine the nonlinear parameters crucial to estimate the resilient modulus of unbound base and subgrade materials.

Material Model Calibration by Deep Learning for Additively Manufactured Alloys

2020 ◽  
Author(s):  
Zihan Wang ◽  
Hongyi Xu ◽  
Yang Li
Keyword(s):  
Machine Learning ◽  
Material Properties ◽  
Model Calibration ◽  
Mesoscale Model ◽  
Input Parameter ◽  
Industrial Applications ◽  
Prediction Errors ◽  
Fish Scale ◽  
Electron Microscopic ◽  
Modeling Framework

Abstract Metal parts manufactured via Powder Bed Fusion (PBF) process show great potential in industrial applications. Hierarchical, heterogeneous microstructure characteristics of the PBF-built alloys pose a significant challenge to the prediction of structural performances. To enable computational engineering of this type of materials, multiscale microstructure modeling framework has been proposed to predict the stochastic material properties. AlSi10Mg built by Selective Laser Melting (SLM) is selected as the demonstrative example. At the microscale, the epitaxial granular structures are reconstructed based on Scanning Electron Microscopic Electron Backscatter Diffraction (SEM EBSD) images. The microscale analysis provides property inputs for the mesoscale model, which captures the fish scale like melt pools at the millimeter scale. The predicted material properties are compared with the experimental data for further calibration of the material constitutive models. One critical challenge is that some parameters in material models cannot be directly obtained from experimental tests. In this work, we establish a machine learning-based model calibration framework to predict the unknown material parameters. Furthermore, several machine learning methods are compared to shed lights on their capability of capturing the relation between input parameter values and the resultant prediction errors.

The development of a machine learning approach to construct a field-scale rock physics transform

Geophysics ◽  
2021 ◽  
pp. 1-55
Author(s):  
Ian Gottschalk ◽  
Rosemary Knight
Keyword(s):  
Machine Learning ◽  
Objective Function ◽  
Material Properties ◽  
Learning Algorithm ◽  
Geophysical Methods ◽  
Rock Physics ◽  
Important Predictor ◽  
Spatial Density ◽  
Sediment Type ◽  
Boosted Decision Trees

The ability to relate geophysical measurements to the material properties of the subsurface is fundamental to the successful application of geophysical methods. Estimating the electrical resistivity from material properties can be challenging at many hydrogeologic field sites, which typically lack the spatial density and resolution of the measurements needed to develop an accurate rock physics relationship. We developed rock physics transforms using the machine learning method of gradient-boosted decision trees (GBDT). We adopted as our study area the coastal Salinas Valley, where saltwater intrusion results in changes in resistivity. We used measurements available in boreholes, including salinity and sediment type, to predict the resistivity. In some transforms, we included as predictors in the GBDT algorithm the location of each measurement and the aquifer corresponding to each measurement. We also explored incorporating the predictions of a baseline rock physics transform as a prior term within the objective function of the GBDT algorithm to guide the predictions made by the machine learning algorithm. The use of location and aquifer information improved the predictions of the GBDT transform by 28% compared to when location and aquifer information were not included. After the salinity, the easting of each measurement was the most important predictor, due to the spatial pattern of salinity changes in the area. The next most important predictor was the aquifer corresponding to each measurement. The benefit of including the baseline transform in the objective function was greatest for small datasets and when the accuracy of the baseline transform was already high. Finally, using the resistivity predicted by the GBDT, we generated 1-D resistivity models, which we used to simulate the acquisition of airborne electromagnetic (AEM) data. In most cases, the 1-D resistivity models and corresponding AEM data matched well with the models and data corresponding to the resistivity measured in boreholes.

scholarly journals Machine learning material properties from the periodic table using convolutional neural networks

2018 ◽  
Vol 9 (44) ◽  
pp. 8426-8432 ◽  
Author(s):  
Xiaolong Zheng ◽  
Peng Zheng ◽  
Rui-Zhi Zhang
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Enthalpy Of Formation ◽  
Convolutional Neural Networks ◽  
Material Properties ◽  
Periodic Table ◽  
Chemical Information ◽  
Learning Material ◽  
The Enthalpy Of Formation

Convolutional neural networks directly learned chemical information from the periodic table to predict the enthalpy of formation and compound stability.

Machine Learning Potential: A Combination of Reactive Force Field and Message Passing Neural Networks

2021 ◽  
Author(s):  
Liyuan Xue ◽  
Feng Guo ◽  
Yushi Wen ◽  
Shiquan Feng ◽  
Xiaona Huang ◽  
...  
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Force Field ◽  
Material Properties ◽  
Message Passing ◽  
Field Model ◽  
Reactive Force ◽  
Computational Tool ◽  
Learning Potential ◽  
Reactive Force Field

The Reactive Force Field (ReaxFF) is a powerful computational tool for exploring material properties. In this work, we proposed an enhanced reactive force field model, which using Message Passing Neural...

scholarly journals Structure of an ultrathin oxide film solved by machine learning enhanced global optimization

2021 ◽  
Author(s):  
Lindsay Richard Merte ◽  
Malthe Kjær Bisbo ◽  
Igor Sokolović ◽  
Martin Setvín ◽  
Benjamin Hagman ◽  
...  
Keyword(s):  
Machine Learning ◽  
Global Optimization ◽  
Material Properties ◽  
Experimental Methods ◽  
Solid Surfaces ◽  
Model Generation ◽  
Diverse Population ◽  
Energy Estimation ◽  
Ultrathin Oxide

Determination of the atomic structure of solid surfaces is a challenge that has resisted solution despite advancements in experimental methods. Theory-based global optimization has the potential to revolutionize the field by providing reliable structure models as the basis for interpretation of experiments and for prediction of material properties. So far, however, the approach has been limited by the combinatorial complexity and computational expense of sufficiently accurate energy estimation for surfaces. We demonstrate how an evolutionary algorithm, utilizing machine learning for accelerated energy estimation and diverse population generation, can be used to solve an unknown surface structure—the (4 x 4) surface oxide on Pt3Sn(111)--based on limited experimental input. The algorithm is efficient and robust, and should be broadly applicable in surface studies, where it can replace manual, intuition based model generation.

scholarly journals Reliable and explainable machine-learning methods for accelerated material discovery

2019 ◽  
Vol 5 (1) ◽  
Author(s):  
Bhavya Kailkhura ◽  
Brian Gallagher ◽  
Sookyung Kim ◽  
Anna Hiszpanski ◽  
T. Yong-Jin Han
Keyword(s):  
Machine Learning ◽  
Material Properties ◽  
Materials Science ◽  
Imbalanced Data ◽  
Predictive Performance ◽  
Confidence Score ◽  
General Purpose ◽  
Performance Loss ◽  
Commercial Applications ◽  
Evaluation Metric

AbstractDespite ML’s impressive performance in commercial applications, several unique challenges exist when applying ML in materials science applications. In such a context, the contributions of this work are twofold. First, we identify common pitfalls of existing ML techniques when learning from underrepresented/imbalanced material data. Specifically, we show that with imbalanced data, standard methods for assessing quality of ML models break down and lead to misleading conclusions. Furthermore, we find that the model’s own confidence score cannot be trusted and model introspection methods (using simpler models) do not help as they result in loss of predictive performance (reliability-explainability trade-off). Second, to overcome these challenges, we propose a general-purpose explainable and reliable machine-learning framework. Specifically, we propose a generic pipeline that employs an ensemble of simpler models to reliably predict material properties. We also propose a transfer learning technique and show that the performance loss due to models’ simplicity can be overcome by exploiting correlations among different material properties. A new evaluation metric and a trust score to better quantify the confidence in the predictions are also proposed. To improve the interpretability, we add a rationale generator component to our framework which provides both model-level and decision-level explanations. Finally, we demonstrate the versatility of our technique on two applications: (1) predicting properties of crystalline compounds and (2) identifying potentially stable solar cell materials. We also point to some outstanding issues yet to be resolved for a successful application of ML in material science.

In silico high throughput screening of bimetallic and single atom alloys using machine learning and ab initio microkinetic modelling

2020 ◽  
Vol 8 (1) ◽  
pp. 107-123 ◽  
Author(s):  
Shivam Saxena ◽  
Tuhin Suvra Khan ◽  
Fatima Jalid ◽  
Manojkumar Ramteke ◽  
M. Ali Haider
Keyword(s):  
Machine Learning ◽  
Ab Initio ◽  
High Throughput ◽  
Material Properties ◽  
In Silico ◽  
High Throughput Screening ◽  
Chemical Engineering ◽  
Materials Science ◽  
Single Atom ◽  
Single Atom Alloys

The advent of machine learning (ML) techniques in solving problems related to materials science and chemical engineering is driving expectations to give faster predictions of material properties.

Sign in / Sign up

Export Citation Format

Share Document