Machine Learning‐Driven Discovery of Thermoset Shape Memory Polymers With High Glass Transition Temperature Using Variational Autoencoders
Problem
Discovering shape memory polymers with high glass transition temperatures for high-temperature uses such as geothermal energy, oil and gas, and aerospace.
Why it matters
High-temperature applications require materials that keep shape memory behavior under extreme conditions.
What the AI did
A machine learning framework combining transfer learning with variational autoencoders explored the polymer chemical design space to identify high-Tg candidates, and an SVM was trained to predict Tg.
Results
SVM achieved R² exceeding 0.87 and mean absolute percentage error as low as 6.43% on the test set.
Impact
New high-Tg SMP candidates were identified computationally; experimental confirmation is not stated in the abstract.
Record
| Venue | Journal of Polymer Science |
|---|---|
| Published | 2024 |
| Work type | article |
| Authors | Amir Teimouri, Guoqiang Li |
| Citations | 10 (OpenAlex, at import time) |
| References | 58 |
| Model or method | Variational autoencoder with transfer learning; SVM predictor |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1002/pol.20241095 |
| Open access | bronze |
| Keywords | shape memory polymers, variational autoencoder, transfer learning, glass transition temperature, materials discovery |
Importance rating
significant — thousands to millions affected
Abstract calls high-Tg SMPs of paramount importance across geothermal, oil and gas, and aerospace; however, only computational candidates identified, no experimental confirmation stated.
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
|