Optimization of RF to alloy elastic modulus prediction based on cuckoo algorithm

Abstract

In recent years, there has been an explosive rise in the combination of density-functional theory (DFT) computation and machine learning in materials science research. In this paper, a cuckoo algorithm-optimized random forest (CS-RF) based alloy elastic modulus prediction model is proposed using density-functional theory. 320 alloy samples are used as test materials to calculate elastic modulus using the density-functional theory method, and the calculated raw data are normalized to establish a cuckoo algorithm-optimized random-forest model to predict alloy elastic modulus. The experiments show that for predicting the elastic modulus of alloys, CS-RF also has obvious advantages in computational efficiency with improved prediction accuracy. In conclusion, CS-RF is an advanced machine learning (ML) method for predicting the elastic modulus of alloys, which can provide a scientific reference for metal elastic modulus prediction.