Engineering Commons^™

Prediction Of Isotropic Strain Hardening Material Properties Using Gradient Boosted Regression Tree Method And Hyperparameter Optimization, Darren R. Promer

Masters Theses

In this study a new technique is proposed for the determination of elastic-plastic stress-strain relations for isotropic materials using the force-displacement output from an instrumented indentation test, finite element simulation, and machine learning methods. This non-destructive testing methodology has promising potential for industry implementation. Applications to benefit from this method include the characterization of localized material properties on surface engineered components, and the post-manufacturing assessment of material properties for items such as additively manufactured metallic components and load-bearing welded joints.

Currently, the capability of the instrumented indentation test for determining material properties is limited to the elastic modulus and surface …