| dc.description.abstract | The present thesis involves research on the ballistic penetration and perforation of steel plates, where the overall goal is to increase the physical understanding of the different phenomena taking place during this type of structural impact. The impact problem considered is perforation of an intermediate thick target plate of high strength steel in the sub-ordnance velocity regime using a hardened steel projectile with a constant mass.
The objective was achieved through material tests and material modelling (using three constitutive relations and two fracture criteria), penetration tests, metallurgical studies and non-linear FEM-analyses. The major concern was the ballistic limit velocity, which indicates the target’s ability to withstand impact of a projectile.
This thesis describes an experimental and a numerical study of the plate perforation problem, where the target strength, nose shape of the projectile and the impact velocity are varied. Three target strengths were considered, using three alloys from one high-strength steel group: Weldox 460 E, Weldox 700 E and Weldox 900 E. The nose shapes of the projectile were blunt, conical and ogival. The influence of constitutive relations and failure criteria for the specific penetration problem has been studied in detail using the non-linear FEM-code LS-DYNA.
Results indicate that the physical mechanisms in the perforation process were in general well described through numerical analyses. Experimental evidence revealed that the ballistic limit velocity increases for increasing target strength in penetration by conical and ogival projectiles. The observation is reasonable, since the projectiles fail by ductile hole enlargement, i.e. the material in front of the moving projectile is pushed aside and the projectile requires more work to perforate a material of higher strength. This same trend was found in the numerical analyses. For blunt projectiles, the projectiles fail by plugging, and the ballistic limit velocity decreases for increasing strength. This behaviour could be explained through a metallurgical examination. It was found that strongly localized shear bands occur during impact. It was further seen that the volume of the shear zone decreases with increasing material strength, which means that the projectile requires less kinetic energy to perforate. Although the plugging process was numerically well predicted for blunt projectiles, the decrease in ballistic limit velocity for increase in target strength could not be predicted in the FEM-analyses regardless of chosen material models. Numerical analyses of this problem, show strong mesh-dependency due to the shear localisation. It is believed that the ballistic limit velocity will eventually converge towards a limit for an element size of the shear band’s width. The available computer capacity became an obstacle in order to model such fine mesh. However, a preliminary study on using a nonlocal approach for the damage parameter in the failure criteria, indicate that there may be a possibility to relieve the mesh-dependency problem. | nb_NO |
