Machine learning has been widely applied to phase prediction and property evaluation in multi-principal element alloys. In this work, a data-driven machine learning framework is proposed to predict the ultimate tensile strength (UTS) and total elongation (TE) of Fe-Co-Cr-Ni-Mn-Al-Ti multi-principal element alloys (MPEAs), offering a cost-effective route for the design of new MPEAs. A dataset was compiled through an extensive literature survey, and six different machine learning models were benchmarked, from which XGBoost was ultimately selected as the optimal model. The feature set was constructed on the basis of theoretical considerations and experimental data reported in the literature, and SHAP analysis was employed to further elucidate the relative importance of individual features. By imposing constraints on the screened features, two alloys predicted to exhibit superior performance under different heat-treatment conditions were identified and fabricated for experimental validation. The experimental results confirmed the reliability of the model in predicting fracture strength, and the errors observed in ductility prediction were critically examined and discussed. Moreover, the strengthening mechanisms of the designed MPEAs were further explored in terms of microstructural characteristics and lattice distortion effects. The alloy design methodology developed in this study not only provides a theoretical basis for exploring unexplored compositional spaces and processing conditions in multi-principal element alloys, but also offers an effective tool for developing novel alloys that simultaneously achieve high strength and good ductility.
In recent years, many studies have focused on modeling the prediction of microstructure and mechanical properties. Traditional methods mainly rely on physical models and empirical formulas, typically requiring extensive experimental data and struggling to describe complex microstructural characteristics accurately. This paper proposes a deep learning model based on an FPN (Feature Pyramid Network) and CNN (Convolutional Neural Network) for extracting features from microstructural images of Fe-C-Mn-Al-Cr steel processed by vacuum smelting and subjected to eight different heat treatment temperatures to predict its mechanical tensile properties. The experimental results show that the EfficientNetFPN model achieves R2 values of 0.96, 0.97, and 0.97 in predicting tensile strength (UTS), yield strength (YS), and elongation at break (TEL), respectively, outperforming models such as EfficientNet, ResNet50, ResNet18, and VGG16 in prediction accuracy. Compared to traditional CNN models, this method fully leverages the feature extraction advantages of EfficientNetFPN, enabling more comprehensive utilization of microstructural information and thereby improving prediction accuracy. Through feature visualization from low to high levels, the model’s understanding of microstructural images can be elucidated, and key microstructural features affecting material performance can be identified.