Physical metallurgy-guided machine learning and artificial intelligent design of ultrahigh-strength stainless steel

Abstract

With the development of the materials genome philosophy and data mining methodologies, machine learning (ML) has been widely applied for discovering new materials in various systems including highend steels with improved performance. Although recently, some attempts have been made to incorporate physical features in the ML process, its effects have not been demonstrated and systematically analysed nor experimentally validated with prototype alloys. To address this issue, a physical metallurgy (PM) -guided ML model was developed, wherein intermediate parameters were generated based on original inputs and PM principles, e.g., equilibrium volume fraction (V f ) and driving force (D f ) for precipitation, and these were added to the original dataset vectors as extra dimensions to participate in and guide the ML process. As a result, the ML process becomes more robust when dealing with small datasets by improving the data quality and enriching data information. Therefore, a new material design method is proposed combining PM-guided ML regression, ML classifier and a genetic algorithm (GA). The model was successfully applied to the design of advanced ultrahigh-strength stainless steels using only a small database extracted from the literature. The proposed prototype alloy with a leaner chemistry but better mechanical properties has been produced experimentally and an excellent agreement was obtained for the predicted optimal parameter settings and the final properties. In addition, the present work also clearly demonstrated that implementation of PM parameters can improve the design accuracy and efficiency by eliminating intermediate solutions not obeying PM principles in the ML process. Furthermore, various important factors influencing the generalizability of the ML model are discussed in detail.

A Review of the Serrated-Flow Phenomenon and Its Role in the Deformation Behavior of High-Entropy Alloys

“…Furthermore, perhaps it will be determined whether there exists a minimum required concentration for the serrated flow to occur. Machine-learning methods could also be employed to elucidate the relationship among the additive concentration, microstructure (such as dislocation dynamics), and serrated-flow behavior [306][307][308][309][310]. Consequently, a greater understanding of solute atom-dislocation pinning dynamics in HEAs would be achieved.…”

Prediction and Analysis of Tensile Properties of Austenitic Stainless Steel Using Artificial Neural Network

“…Here, the samples that are water-quenched in the database are set as the label 1 and the air-cooled ones without water quenching are set as the label 0. These features are related to the physical metallurgy which is important for modeling and property predictions of ASS [37]. There are some other features in the original database, such as the type of melting, grain size and the form of products, but the data of them are incomplete or they have a lower correlation with tensile properties.…”

Cluster-formula-embedded machine learning for design of multicomponent β-Ti alloys with low Young’s modulus

“…respectively, where n is the number of samples, y is the mean Young's modulus of samples, y i and f(x i ) are the experimental Young's modulus and the predicted value by ML, respectively. Since the dataset capacity of 82 samples is somewhat limited in the present work, it is necessary to employ the multiple hold-out method to calculate the RMSE and R 2 of ML models for the guarantee of accuracy 31 . This procedure was repeated for 500 times in these three models by partitioning the training set and testing set randomly with a ratio of 9/1, and then the RMSE and R 2 values were counted, as shown in Fig.…”

“…Similarly, the combination of structural and compositional features (cohesive energy, atomic radius, and electronegativity) with the ML could well predict the elastic properties of Al-Co-Cr-Fe-Ni high-entropy alloys 30 . Xu et al embedded the physical-metallurgy parameters, the volume fraction, and driving force of second-phase precipitates that represent the microstructural features, into the ML model to design advanced ultrahighstrength stainless steels successfully 31 . The optimal prototype alloy predicted by this model was well demonstrated by experiments, in which an excellent agreement could be obtained for the predicted optimal parameter settings and the final mechanical property (microhardness).…”