.png){kind=icon}
.png){kind=icon}
Return
Machine learning enabled processing map generation for high-entropy alloy
.png){kind=icon}
PRE
AI
DOI：10.1016/j.scriptamat.2023.115543.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Identifying optimum processing conditions is necessary for new material development. The flow curves can be used to develop the processing map for an alloy. The current study trained multiple machine learning models such as Random Forest Regressor (RFR), K Nearest Neighbors (KNN), Extra Tree Regressor (ETR) and Artiflcial Neural Network (ANN) to predict the flow behaviour of the material. The testing R2 flt score of more than 0.99 was obtained for all four algorithms, and trained models were used to generate the flow curves at various temperature strain rate combinations for CoCrFeNiTa0.395 eutectic high entropy alloy. A processing map was developed using the results from ANN and validated with the experimental microstructure observations.
Keywords:
Processing maps
Eutectic high entropy alloys
Machine learning
Hot deformation
Journal
IF:
5.6
Papers:
1.6W
Citations:
5.1W