Understanding the role of segmentation on process-structure–property predictions made via machine learning

Massey, Caroline; Saldaña, Christopher; Moore, David G.

doi:10.1007/s00170-022-09003-8

Cited by 3 publications

(1 citation statement)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…ML approaches are gaining popularity in the age of big data, especially in materials science . Researchers used ML models in various aspects of materials science including material properties prediction, − structural topology optimization, − design of spinodoid metamaterials, multi-scale modeling of porous media, design of tessellate composites, − architected materials design, material property extraction, prediction of hyperelastic or plastic behaviors, − process–structure–property linkage, defect identification, and so on. One of the major goals of ML models is to accomplish high-throughput measurement of critical characteristics of materials under different conditions .…”

Section: Introductionmentioning

confidence: 99%

Data-Driven Machine Learning to Predict Mechanical Properties of Monolayer Transition-Metal Dichalcogenides for Applications in Flexible Electronics

Malakar

Thakur

Nahid

et al. 2022

ACS Appl. Nano Mater.

View full text Add to dashboard Cite

Understanding of the material properties of layered transition-metal dichalcogenides (TMDs) is critical for their applications in flexible electronics. Data-driven machine learning (ML)-based approaches are being developed in contrast to the traditional experimental or computational methods to predict and understand material properties under varied operating conditions. In this study, we used two ML algorithms, namely, long short-term memory (LSTM) and feed forward neural network (FFNN), combined with molecular dynamics (MD) simulations to predict the mechanical properties of MX 2 (M = Mo, W and X = S, Se) TMDs. The LSTM model is found to be capable of predicting the entire stress−strain response, whereas the FFNN is used to predict material properties such as fracture stress, fracture strain, and Young's modulus. The effects of operating temperature, chiral orientation, and pre-existing crack size on the mechanical properties are thoroughly investigated. We carried out 1440 MD simulations to produce the input dataset for the neural network models. Our results indicate that both LSTM and FFNN are capable of predicting the mechanical response of monolayer TMDs under different conditions with more than 95% accuracy. The FFNN model exhibits lower computational cost than LSTM; however, the capability of the LSTM model to predict the entire stress−strain curve is advantageous for assessing material properties. The study paves the pathway toward extending this approach to predict other important properties, such as optical, electrical, and magnetic properties of TMDs.

show abstract