A machine learning aided interpretable model for rupture strength prediction in Fe-based martensitic and austenitic alloys

Abstract

The class of 9–12% Cr ferritic-martensitic alloys (FMA) and austenitic stainless steels have received considerable attention due to their numerous applications in high temperature power generation industries. To design high strength steels with prolonged service life requires a thorough understanding of the long-term properties, e.g., creep rupture strength, rupture life, etc., as a function of the chemical composition and processing parameters that govern the microstructural characteristics. In this article, the creep rupture strength of both 9–12% Cr FMA and austenitic stainless steel has been parameterized using curated experimental datasets with a gradient boosting machine. The trained model has been cross validated against unseen test data and achieved high predictive performance in terms of correlation coefficient ($$R^{2} > 0.98 $$ R 2 > 0.98 for 9–12% Cr FMA and $$R^{2} > 0.95 $$ R 2 > 0.95 for austenitic stainless steel) thus bypassing the need for additional comprehensive tensile test campaigns or physical theoretical calculations. Furthermore, the feature importance has been computed using the Shapley value analysis to understand the complex interplay of different features.

Machine learning assisted prediction of the Young’s modulus of compositionally complex alloys

“…In our data sets, tree-based ensemble type models perform better than other models to predict Young's modulus. Ensemble type algorithm showed better performance in other studies to predict materials properties 34,48,49 . Ensemble methods are meta algorithms that combine several base models to produce a better predictive model.…”

DEM- and GIS-Based Analysis of Soil Erosion Depth Using Machine Learning

“…Friedman [43] proposed the gradient boosting machine as a simple and highly flexible machine learning tool. It is a widely used machine learning algorithm that has been shown to be effective in a variety of applications [44][45][46]. The basic idea behind GBM is to build a prediction model using a set of poor learning algorithms, most commonly decision trees.…”

Uncertainty quantification for Bayesian active learning in rupture life prediction of ferritic steels

“…The data used in this study were collected over 30 years of concerted efforts from government and non-government initiatives 16 . The overarching goal of probabilistic machine learning is to make reliable predictions for unknown alloys as well as to accelerate the data collection process to improve the model performance.…”

“…Neural Network has been successfully employed in the past by Cole et al 12 and Brun et al 13 to model the creep rupture strength from physical and processing parameters. In a recent article, it was shown that a Gradient Boosting Algorithm 14 can be used to efficiently predict the rupture life 15 and rupture strength 16 of 9–12 wt% Cr ferritic-martensitic steels and austenitic stainless-steels. However, oftentimes the prediction is unreliable for low confidence modeling, where the data is either scarce and highly non-linear or the uncertainty associated with the prediction is not available.…”