Machine learning approach for predicting yield strength of nitrogen-doped CoCrFeMnNi high entropy alloys at selective thermomechanical processing conditions
Mokali Veeresham, Reliance Jain, Unhae Lee, Nokeun Park
Abstract
The yield strength property is important to consider while designing high entropy alloys (HEAs). In order to obtain the desired yield strength of HEAs through the experimental method it is a difficult, expensive, and time-consuming process due to the broad composition space available. The room temperature yield strength property of nitrogen-doped (CoCrFeMnNi)100-x-Nx HEAs at preferred thermomechanical conditions was predicted using the machine learning (ML) technique based on the linear regression model in the present investigation. The yield strength prediction result of 2% nitrogen-doped CoCrFeMnNi HEA subsequently cold-rolled 91 (%) and annealed at 850 °C temperature consisting of 563.6 MPa is consistent with the experimental value of 556 MPa. It implies that the yield strength predictions of (CoCrFeMnNi)100-x-Nx HEAs are accurate. As a result, selecting suitable models and material parameters to design a wide range of materials with superior properties attributed to various compositions of HEAs through ML technology could be a potential approach.