About this Abstract |
Meeting |
2022 TMS Annual Meeting & Exhibition
|
Symposium
|
Dynamic Behavior of Materials IX
|
Presentation Title |
Multi-fidelity Machine Learning Based Approach to Predict Local Strain Response |
Author(s) |
Tyler Dillard, Nolan Lewis, Abhijeet Dhiman, Vikas Tomar |
On-Site Speaker (Planned) |
Tyler Dillard |
Abstract Scope |
High strain dynamic response of materials is of key importance for material development. Of particular interest is modeling and predicting the wave propagation within a material in response to high impact forces. PMMA(Polymethyl Methacrylate) samples were subjected to flat nail impacts at high velocities using an in-house built table-top gas gun. Sensors (force sensitivity resistors) were used to measure the material load response of examined samples. Digital Image Correlation (DIC) was used with a high-speed camera to record the local surface deformation and strain field. The use of an FEM simulation enabled the ability to produce large amounts of high through-put data to produce a data set for a low fidelity model. The model predicts the strain profile (DIC data) of the target material (PMMA) during dynamic impact. Composite neural-net architecture is implemented to model a multi-fidelity approach and comparisons were made between low-fidelity, high-fidelity, and multi-fidelity predictions. |
Proceedings Inclusion? |
Planned: |
Keywords |
Machine Learning, Mechanical Properties, Modeling and Simulation |