Multi-objective materials bayesian optimization with active learning of design constraints: Design of ductile refractory multi-principal-element alloys

Bayesian Optimization (BO) has emerged as a powerful framework to efficiently explore and exploit ma-terials design spaces. To date, most BO approaches to materials design have focused on the materials discovery problem as if it were a single expensive-to-query 'black box' in which the target is to optimize a single objective (i.e., material property or performance metric). Also, such approaches tend to be con-straint agnostic. Here, we present a novel multi-information BO framework capable of actively learning materials design as a multiple objectives and constraints problem. We demonstrate this framework by op-timally exploring a Refractory Multi-Principal-Element Alloy (MPEA) space, here specifically, the system Mo-Nb-Ti-V-W. The MPEAs are explored to optimize two density-functional theory (DFT) derived ductility indicators (Pugh's Ratio and Cauchy pressure) while learning design constraints relevant to the manufac-turing of high-temperature gas-turbine components. Alloys in the BO Pareto-front are analyzed using DFT to gain an insight into fundamental atomic and electronic underpinning for their superior performance, as evaluated within this framework. (c) 2022 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

Automated summary

This paper presents a novel multi-information BO framework for learning materials design as a multiple objectives and constraints problem. By optimizing ductility indicators and studying manufacturing constraints, the framework efficiently explores a multi-principal-element alloy space.