Domain knowledge-guided interpretive machine learning: formula discovery for the oxidation behavior of ferritic-martensitic steels in supercritical water

Abstract

A general formula with high generalization and accurate prediction power is highly desirable for science, technology and engineering. In addition to human beings, artificial intelligence algorithms show great promise for the discovery of formulas. In this study, we propose a domain knowledge-guided interpretive machine learning strategy and demonstrate it by studying the oxidation behavior of ferritic-martensitic steels in supercritical water. The oxidation Cr equivalent is, for the first time, proposed in the present work to represent all contributions of alloying elements to oxidation, derived by our domain knowledge and interpretive machine learning algorithms. An open-source tree classifier for linear regression algorithm is also, for the first time, developed to materialize the formula with collected data. This algorithm effectively captures the linear correlation between compositions, testing environments and oxidation behaviors from the data. The sure independence screening and sparsifying operator algorithm finally assembles the information derived from the tree classifier for linear regression algorithm, resulting in a general formula. The general formula with the determined parameters has the power to predict, quantitatively and accurately, the oxidation behavior of ferritic-martensitic steels with multiple alloying elements exposed to various supercritical water environments, thereby providing guidance for the design of anti-oxidation steels and hence promoting the development of power plants with improved safety. The present work demonstrates the power of domain knowledge-guided interpretive machine learning with respect to the data-driven discovery of physics-informed formulas and the acceleration of materials informatics development.

Achieving High Performance Aqueous Zn-Ion Batteries via Interfacial Coating of N, P Dual-Doped Biomass Porous Carbon on Zn Metal Anode

“…The above results show that the polar functional groups containing N, P, and O atoms favor electrolyte penetration and the fast transfer kinetics of Zn 2+ . 52 This causes a uniform deposition of Zn 2+ and improves the stability of the Zn anode. To reveal the zincophilic performances of N, P dual-doped carbon, the first-principles method was adopted to calculate the adsorption energies of Zn atom.…”

“…To reveal the zincophilic performances of N, P dual-doped carbon, the first-principles method was adopted to calculate the adsorption energies of Zn atom. ,,, The N, P co-doped graphene (named as NP1–G, NP2–G) was constructed in the Figure S6, compared with pristine G (PG) and N-doped G (NG). The adsorption of Zn atom on the described surfaces was investigated, and corresponding adsorption configurations were showed in Figure S7.…”

Design and Selection of High Entropy Alloys for Hardmetal Matrix Applications Using a Coupled Machine Learning and Calculation of Phase Diagrams Methodology

“…A positive/ negative SHAP value indicates a positive/negative impact on model predictions, associated with the likelihood to predict FCC phase formation and values of hardness, both for the classification and regression tasks, respectively. [42,71] Hence, the wider the horizontal distribution of points, the greater the influence of that feature on the models' predictions. Additionally, features are ordered on the y-axis by their importance for predictions.…”

Divide and Conquer: Machine Learning Accelerated Design of Lead-Free Solder Alloys with High Strength and High Ductility

“…When studying the glass-forming ability in bulk amorphous metals, Xiong and Zhang [18] use a data augmentation technique to improve original small and imbalanced data. Cao et al [19] proposed a domain knowledge-guided interpretive machine learning strategy and demonstrated it by studying the oxidation behavior of ferriticmartensitic steels in supercritical water. A ML algorithm of Tree-Classifier for Linear Regression (TCLR) is developed which effectively captures the linear correlation between compositions, testing environments and oxidation behaviors from sparse data with high dimensions.…”

“…Consequently, a generalized Arrhenius oxidation formula is accomplished with very high prediction accuracy and wide generality. Wei et al [54] adopted the same strategy [19] to discover high interpretive formula describing the high temperature oxidation behavior of FeCrAlCoNi-based high entropy alloys (HEAs). The TCLR algorithm was used to extract the spectrums of activation energy Q and time exponent m from the complex and high dimensional feature space, which automatically gives the spectrum of pre-factor.…”