About this Abstract |
| Meeting |
2026 TMS Annual Meeting & Exhibition
|
| Symposium
|
2026 Technical Division Student Poster Contest
|
| Presentation Title |
SPG-89: Machine Learning Models Trained on Discrete Dislocation Dynamics Data for the Interactions Between an Infinite Line and a Prismatic Dislocation Loop |
| Author(s) |
Zeyu Dong, Liyi Zhu |
| On-Site Speaker (Planned) |
Zeyu Dong |
| Abstract Scope |
Understanding dislocation-obstacle interactions is fundamental to predicting the yield strength and hardening behavior of metallic materials, particularly those subjected to irradiation, which generates a high density of prismatic loops.This study aims to train machine learning (ML) models on discrete dislocation dynamics (DDD) data to predict the importance of the initial dislocation configuration affecting the interaction between an infinite line and a prismatic dislocation loop. The critical stress required for an infinite line to bypass a prismatic loop was quantified using the OpenDiS simulation software. A comprehensive dataset was generated by systematically varying key geometric parameters, including the distance between two dislocations, the prismatic loop diameter, and the angle of the prismatic loop. The critical yield stress for each configuration was determined by applying a series of incremental constant forces. ML models were then trained to predict the critical stress from these geometric inputs. |
| Proceedings Inclusion? |
Undecided |