About this Abstract |
Meeting |
2022 TMS Annual Meeting & Exhibition
|
Symposium
|
30 Years of Nanoindentation with the Oliver-Pharr Method and Beyond
|
Presentation Title |
Using Machine Learning Approaches to Enable Insights in Nanoindentation Tip Wear |
Author(s) |
Claus Trost, Stanislav Zak, Sebastian Schaffer, Megan J. Cordill |
On-Site Speaker (Planned) |
Claus Trost |
Abstract Scope |
Extraction of material mechanical behaviour exceeding the classical Oliver and Pharr analysis is a complex task. Therefore, different approaches such as machine learning algorithms are being used to interpret indentation data. In this study, different machine learning methods will be used on simulated 2D, 3D simulations to interpret nanoindentation experiments. The simulated data will be used to find features in indentation curves and train machine learning algorithms to predict both tip wear and material parameters. The direct interpretation of the wear is expected to give new insights in nanoindentation experiments. The machine learning methods will be analysed using the game theory-based model agnostic SHAP (Shapley Additive exPlanations) approach and light will be shed on the impact of different features on the output of the respective model. SHAP is expected to enhance the understanding of machine learning problems in the field of nanoindentation and many other areas. |
Proceedings Inclusion? |
Planned: |
Keywords |
Mechanical Properties, Machine Learning, Thin Films and Interfaces |