Constituent-Material-Anchored Continual Learning for Full Stress–Strain Prediction of Multi-Material PETG/PC-ABS MEX Laminates
Ramachandran Avala Subramanian, Mahalingam Nainaragaram Ramasamy, Michal Prauzek, Quoc-Phu Ma, Jaromir Konecny, Ales Sliva · 2026
Read Constituent-Material-Anchored Continual Learning for Full Stress–Strain Prediction of Multi-Material PETG/PC-ABS MEX Laminates by Ramachandran Avala Subramanian, Mahalingam Nainaragaram Ramasamy, Michal Prauzek, Quoc-Phu Ma, Jaromir Konecny, Ales Sliva (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13364054.txt.
Description
Ramachandran Avala Subramanian and colleagues present a 2026 study on predicting the full stretch curve of two-plastic printed laminates with a learning model that updates as new parts are tested. The coupons layer PETG with PC-ABS, two printing plastics, in varied shares. A rule-of-mixtures line from the pure plastics forms the base guess, and a gradient-boosted model learns the corrections from strain, phase, and layer share.
Peak strength alone misled: one design nailed it in a blind guess yet missed the full curve badly. After just one real coupon of each new design was added, curve errors fell by two thirds or more across all five layouts. The authors offer the anchored, continual approach as a cheap screen for printed part designs.
Catalog record
| CREATOR | Ramachandran Avala Subramanian, Mahalingam Nainaragaram Ramasamy, Michal Prauzek, Quoc-Phu Ma, Jaromir Konecny, Ales Sliva |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 63366 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |