Rapid discovery of high hardness multi-principal-element alloys using a generative adversarial network model

Roy, Abhinaba; Hussain, Aqumar; Sharma, Prince; Balasubramanian, Ganesh; Taufique, M. F. N.; Devanathan, Ram; Singh, Prashant; Johnson, Duane D.

doi:10.1016/j.actamat.2023.119177

Cited by 8 publications

(1 citation statement)

References 66 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The developed algorithms were found to be generic and applicable for predicting other mechanical properties as well but limited to a particular system of alloy. Roy et al 13 proposed an algorithmic framework using Generative Adversarial Networks (GANs) to generate novel compositions based on available data of MPEAs with higher hardness values. This is followed by identification of alloy with the highest hardness through a hardness predictor.…”

Section: Introductionmentioning

confidence: 99%

Designing of high entropy alloys with high hardness: a metaheuristic approach

Poonia,

Kishor,

Ayyagari

2024

Sci Rep

View full text Add to dashboard Cite

The near-infinite compositional space of high-entropy-alloys (HEAs) is a huge resource-intensive task for developing exceptional materials. In the present study, an algorithmic framework has been developed to optimize the composition of an alloy with chosen set of elements, aiming to maximize the hardness of the former. The influence of phase on hardness prediction of HEAs was thoroughly examined. This study aims to establish generalized prediction models that aren’t confined by any specific set of elements. We trained the HEA identification model to classify HEAs from non-HEAs, the multi-labeled phase classification model to predict phases of HEAs also considering the processing route involved in the synthesis of the alloy, and the hardness prediction model for predicting hardness and optimizing the composition of the given alloy. The purposed algorithmic framework uses twenty-nine alloy descriptors to compute the composition that demonstrates maximum hardness for the given set of elements along with its phase(s) and a label stating whether it is classified as HEA or not.

show abstract

Section: Introductionmentioning

confidence: 99%