Phase Transformation Temperature Prediction in Steels via Machine Learning
Problem
Predicting the four key steel phase transformation temperatures (Ac1, Ac3, MS, BS) and understanding which alloying elements and atomic parameters influence them, which matters for steel design and heat treatment.
Why it matters
Phase transformation temperatures play an important role in steel design, production, and heat treatment processes.
What the AI did
Used an improved LightGBM gradient-boosting model to predict the four transformation temperatures and rank the influence of alloying elements and eighteen atomic-parameter features.
Results
Training accuracy significantly improved by adding atomic-parameter features; melting temperature, linear thermal expansion coefficient, Waber–Cromer pseudopotential radii, and valence electron number ranked top four of eighteen.
Impact
The abstract claims improved training accuracy and identified influential atomic parameters, without numeric benchmarks.
Record
| Venue | Materials |
|---|---|
| Published | 2024 |
| Work type | article |
| Authors | Yupeng Zhang, Lin Cheng, Aonan Pan, Chengyang Hu, Kaiming Wu |
| Citations | 13 (OpenAlex, at import time) |
| References | 71 |
| Model or method | LightGBM (improved version) |
| Organization | not stated in source |
| Primary source | https://doi.org/10.3390/ma17051117 |
| Open access | gold |
| Keywords | steel, phase transformation temperature, LightGBM, machine learning, alloying elements |
Importance rating
real but narrower advance
Phase transformation temperatures matter for steel design, production, and heat treatment across a large industry, but the work is a modeling study with no stated scale.
More in Engineering & Materials
| # | Paper | Year | Citations | Importance |
|---|---|---|---|---|
| 1 | Scaling deep learning for materials discovery | 2023 | 1,471 |
8
|
| 2 | Application of Machine Learning in Material Synthesis and Property Prediction | 2023 | 165 |
7
|
| 3 | Machine learning-guided discovery of ionic polymer electrolytes for lithium metal batteries | 2023 | 115 |
7
|
| 4 | Machine Learning for Structural Health Monitoring of Aerospace Structures: A Review | 2025 | 44 |
7
|
| 5 | Power System Transient Stability Assessment Based on Machine Learning Algorithms and Grid Topology | 2023 | 42 |
7
|