A5058312061- Machine Learning in Materials Science
- High Entropy Alloys Studies
- Metallic Glasses and Amorphous Alloys
- Phase-change materials and chalcogenides
- Crystallization and Solubility Studies
- Magnetic Properties of Alloys
- Metallurgical Processes and Thermodynamics
- Advanced Materials Characterization Techniques
- Shape Memory Alloy Transformations
- Glass properties and applications
- High Temperature Alloys and Creep
- Additive Manufacturing Materials and Processes
- X-ray Diffraction in Crystallography
- Computational Drug Discovery Methods
- Manufacturing Process and Optimization
- Magnetic Properties and Applications
- Advanced Surface Polishing Techniques
- Metallurgy and Material Forming
- Quasicrystal Structures and Properties
- Chemical Thermodynamics and Molecular Structure
- Thermal and Kinetic Analysis
- Welding Techniques and Residual Stresses
- X-ray Spectroscopy and Fluorescence Analysis
- Microstructure and Mechanical Properties of Steels
Shanghai Jiao Tong University
2021-2024
Chinese National Human Genome Center at Shanghai
2021-2023
Phase boundary indicates the conditions of transition between phase regions, which is a key constituent diagram. We propose an approach to determine and its uncertainty in phase-composition map based on data from high-throughput experimentation. Bayesian logistics regression combined with domain knowledge diagrams was applied approach. For typical ternary isothermal section, both linear nonlinear boundaries as well vertices tie triangle were modeled quantify consideration couple affecting...
Abstract Hierarchical clustering algorithm has been applied to identify the X‐ray diffraction (XRD) patterns from a high‐throughput characterization of combinatorial materials chips. As data quality is usually correlated with acquisition time, it important study hierarchical performance as function in order optimize efficiency experiments. This work investigated effects signal‐to‐noise ratio on using 29 distance metrics for XRD Fe−Co−Ni ternary chip. It found that accuracies evaluated by F1...
Many works have reported machine learning (ML) to predict the glass-forming range (GFR) of metallic glasses. However, datasets used train model were mostly imbalanced and clustered around only a small portion composition space that made hard extrapolate. In this work, generalization deficiency ML was addressed by combining combinatorial materials chip (CMC) high-throughput (HiTp) experimentation modelling, importance comprehensive data highlighted. By training models with HiTp dataset...
There is a gap between the cooling rate of thin film deposition (~108 K/s) and bulk metallic glass (BMG) casting (~103 K/s). In this work, we tried to corelate glass-forming range (GFR) determined from combinatorial materials chip (CMC) with ability (GFA) BMG. Data 5 CMCs reported critical diameter (Dmax) BMGs were used as dataset. It found that full-width at half maximum (FWHM) first sharp diffraction peak (FSDP) good indicator GFA BMG suggested by Liu et al. [1]. Strong positive...
The Fe–Ni based classical Invar alloy and the Fe–Ni–Co super are basis for designing multicomponent new alloys with low thermal expansion. In this work, relationships among expansion, chemical composition, crystal structure magnetism in system were systematically studied by high-throughput characterization of combinatorial materials chips (CMC). distribution coefficients expansion (CTE, 80~200°C) measured in-situ micro-beam X-ray diffraction on CMC showed CTE regions consistent reference...
Vegard’s law factor (VLF) and volume size (SF) of binary substitutional metallic solid solutions (BSMSS) are crucial in alloy design. However, general models for accurately predicting these structure parameters still undeveloped. In this work, a kernel ridge regression (KRR) model was developed to predict VLF SF BSMSS based on 418 published entries elemental descriptors. The KRR achieves an average R2 score 0.62, Pearson coefficient 0.80 0.72, 0.87 among 100 random splits training/testing...
Lack of experimental data, inconsistence the dataset used as input and complexity model caused by high dimensionality are all problems lies in construction multicomponent phase diagram. In this context, developing new kind approach employed for rapid diagrams is essential. Here, a method based on combination combinatorial material science Bayesian logistics regression developed to construct diagram under principle Gibbs rule. Relationships among factors such data point density, shape...
In this paper, the high-throughput (HiTp) combinatorial materials chip method was used to generate systematic and balanced data on glass-forming range (GFR) of some ternary systems. Thirteen chips covering full composition investigated systems were prepared characterized for training validation purposes, among which GFRs FeZrHf, CoTiHf, CoZrHf, FeCoHf, FeTiHf, FeTiZr, FeZrV ternaries have been newly found. By a random forest (RF) machine learning (ML) model using several combinations between...