Machine learning assisted modelling and design of solid solution hardened high entropy alloys

Huang, Xiaofeng; Jin, Chunyuan; Zhang, Chi; Zhang, Hu; Fu, Hanwei

doi:10.1016/j.matdes.2021.110177

Cited by 54 publications

(30 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While the BD method identifies mean_NValance as important, the SHAP identifies mean_CovalentRadius as dominant. Both outcomes are consistent with previous published reports in the literature 44 , 48 – 53 . Despite exhibiting similar patterns (as visualized in Fig.…”

Section: Resultssupporting

confidence: 93%

A comparison of explainable artificial intelligence methods in the phase classification of multi-principal element alloys

Lee

Ayyasamy

et al. 2022

Sci Rep

View full text Add to dashboard Cite

We demonstrate the capabilities of two model-agnostic local post-hoc model interpretability methods, namely breakDown (BD) and shapley (SHAP), to explain the predictions of a black-box classification learning model that establishes a quantitative relationship between chemical composition and multi-principal element alloys (MPEA) phase formation. We trained an ensemble of support vector machines using a dataset with 1,821 instances, 12 features with low pair-wise correlation, and seven phase labels. Feature contributions to the model prediction are computed by BD and SHAP for each composition. The resulting BD and SHAP transformed data are then used as inputs to identify similar composition groups using k-means clustering. Explanation-of-clusters by features reveal that the results from SHAP agree more closely with the literature. Visualization of compositions within a cluster using Ceteris-Paribus (CP) profile plots show the functional dependencies between the feature values and predicted response. Despite the differences between BD and SHAP in variable attribution, only minor changes were observed in the CP profile plots. Explanation-of-clusters by examples show that the clusters that share a common phase label contain similar compositions, which clarifies the similar-looking CP profile trends. Two plausible reasons are identified to describe this observation: (1) In the limits of a dataset with independent and non-interacting features, BD and SHAP show promise in recognizing MPEA composition clusters with similar phase labels. (2) There is more than one explanation for the MPEA phase formation rules with respect to the set of features considered in this work.

show abstract

Section: Resultssupporting

confidence: 93%

A comparison of explainable artificial intelligence methods in the phase classification of multi-principal element alloys

Lee

Ayyasamy

et al. 2022

Sci Rep

View full text Add to dashboard Cite

show abstract

“…Previous work has also described the significance and importance of VEC in physical characteristics and phase stability in alloys [ 53 , 54 , 64 , 65 ]. Similar results were found by Huang et al [ 66 ] about the VEC , which holds the top position in features importance while predicting the hardness of RHEAs using the RF model. They also highlight that VEC is important in charge transfer, which impacts RHEA strength because of lattice distortion.…”

Section: Resultssupporting

confidence: 89%

Predicting Elastic Constants of Refractory Complex Concentrated Alloys Using Machine Learning Approach

et al. 2022

View full text Add to dashboard Cite

Refractory complex concentrated alloys (RCCAs) have drawn increasing attention recently owing to their balanced mechanical properties, including excellent creep resistance, ductility, and oxidation resistance. The mechanical and thermal properties of RCCAs are directly linked with the elastic constants. However, it is time consuming and expensive to obtain the elastic constants of RCCAs with conventional trial-and-error experiments. The elastic constants of RCCAs are predicted using a combination of density functional theory simulation data and machine learning (ML) algorithms in this study. The elastic constants of several RCCAs are predicted using the random forest regressor, gradient boosting regressor (GBR), and XGBoost regression models. Based on performance metrics R-squared, mean average error and root mean square error, the GBR model was found to be most promising in predicting the elastic constant of RCCAs among the three ML models. Additionally, GBR model accuracy was verified using the other four RHEAs dataset which was never seen by the GBR model, and reasonable agreements between ML prediction and available results were found. The present findings show that the GBR model can be used to predict the elastic constant of new RHEAs more accurately without performing any expensive computational and experimental work.

show abstract

“…Huang et al [36] also utilized ML to model SSS in HEA using atomic environment and interactions as the input parameters. The developed ML model was used to predict hardness of single solid solution HEAs by combining the critical theoretical SSS descriptors such as shear modulus, atomic size mismatch, bulk modulus and phase stability descriptors such as mixing enthalpy, mixing entropy, and a phase stability descriptor, Pauling electronegativities, valence electron concentration (VEC) and melting temperature.…”

Section: Machine Learningmentioning

confidence: 99%

“…local chemical environment, Huang et al [36] proposed a modified physical model for SSS by adding the charge transfer (dQ) between atoms to the Varvenne and Maresca-Curtin models. This charge transfer induces atomic-level pressure (DP solute ) fluctuations, which dominates the SSS of several Cantor based HEAs [35].…”

Section: Machine Learningmentioning

confidence: 99%

Solid Solution Strengthening in High-Entropy Alloys

Ondicho¹,

Alunda²,

Kamau³

2023

High Entropy Materials - Microstructures and Properties

View full text Add to dashboard Cite

This book chapter discusses solid solution strengthening (SSS) as one of the main hardening mechanisms in high-entropy alloys (HEAs) that form basis as one of its core effects (lattice distortion). The various techniques used to quantify SSS and the role of different substitutional and interstitial elements/atoms in improving the strength of HEAs are outlined in detail. This review provides a good assessment on ways to enhance the mechanical properties of HEAs to suit the extreme demands of modern engineering applications. Based on theoretical modeling and experimental validation, Al and Nb provide superior substitutional SSS in face-centered cubic and body-centered cubic crystal structures, while carbon has a 50% more effect on improving the mechanical properties of HEAs than in stainless and twinning-induced plasticity steels. Moreover, a detailed description of the application of machine learning in design of HEAs shows that trial and error can eliminated in identifying HEAs with exceptional yield strength. The atomic size difference should be used to evaluate the lattice distortion effect.

show abstract

Machine learning assisted modelling and design of solid solution hardened high entropy alloys

Cited by 54 publications

References 46 publications

A comparison of explainable artificial intelligence methods in the phase classification of multi-principal element alloys

A comparison of explainable artificial intelligence methods in the phase classification of multi-principal element alloys

Predicting Elastic Constants of Refractory Complex Concentrated Alloys Using Machine Learning Approach

Solid Solution Strengthening in High-Entropy Alloys

Contact Info

Product

Resources

About