<p> It is important to understand the dynamic failure behavior of structures subjected to impact loading in order to improve the survivability. Materials under impact are utterly affected by large deformations, high strain-rates, temperature softening and varying stress-states, which finally may lead to failure. It is shown that the impact characteristics are prone to change with several independent factors such as; impact speed, material thickness, and shape and orientation of the impacting object. Validated numerical simulations of impact tests reveal that the failure on ductile metals occur at certain locations of the failure locus that is constructed on a space as a function of all three stress invariants, which indicates that the failure depends profoundly on the state-of-stress. It is shown that existing material models are not always successful enough to cover the whole range of the failure locus and predict the failure. Therefore, it is a common practice to use different sets of material model parameters tuned or calibrated to cover a specific region of the failure loci in an ad hoc manner for practical reasons to match particular test results. Even in that case, specially tuned material properties are not capable of predicting these limited cases if differences in the mesh size and pattern need to be considered. </p><p> In this dissertation a new, generic, thermo-elastic/viscoplastic material model with regularized failure is introduced. The new material model is implemented into a non-linear, explicit dynamics finite element code, LS-DYNA. A von Mises type isotropic, isochoric plasticity is utilized, where isotropic hardening, strain-rate hardening and temperature softening is considered. The model takes adiabatic heating and softening into account due to the plastic work. The constitutive relation is coupled with a new regularized accumulated failure law that is specifically developed to cover a large extent of the failure locus as a function of state-of-stress, strain-rate and temperature. Regularization treatment is implemented to reduce mesh size dependency especially for the problems where softening and failure is involved for the failure prediction. </p><p> Ductile deformation and failure mechanism of 2024-T3/T351 aluminum alloy is investigated experimentally and numerically for quasi-static and dynamic conditions at various temperatures and stress-states. An intelligently contrived test matrix is developed by designing specific test specimens with different geometries that can construct a failure locus as a function of state-of-stress, strain-rate and temperature. An inverse material characterization algorithm is then introduced to generate input data for the new material model. Tabulated inputs of characterized material test results are directly used for both the constitutive and failure treatment of the new material model. Component based specimen tests that are used to characterize the material input properties and full-scale impact tests that are performed at different target thicknesses and impact speeds are used to validate and show the robustness, accuracy and efficiency of the new material model. </p><p> It is shown that the new material model is capable of predicting ballistic limit and failure modes accurately for structures under impact even if the failure mode changes drastically. It is also shown that the new regularization model provides less mesh size dependency. These associated features of the model suggest that the new material model can be used as a promising generic tool for diverse applications of dynamic ductile deformation and failure phenomenon. </p>
| Identifer | oai:union.ndltd.org:PROQUEST/oai:pqdtoai.proquest.com:3597748 |
| Date | 27 November 2013 |
| Creators | Buyuk, Murat |
| Publisher | The George Washington University |
| Source Sets | ProQuest.com |
| Language | English |
| Detected Language | English |
| Type | thesis |
Page generated in 0.0019 seconds