Structural Response in Ship-Platform and Ship-Ice Collisions

Collision events may have severe consequences, and it is important to design both ship and offshore structures so that they have sufficient resistance to a collision event. The main purpose of the work herein is to improve the methods for description of the material behavior in nonlinear finite element analysis (NLFEA), and to study the physics of the collision process through numerical simulations of both experiments and full-scale scenarios. Simplified methods are proposed based on the findings. The shape of the stress-strain curve of the steel material is found to have a large effect on how strains localize in a deformation process, and thereby on the fracture initiation and propagation. Strain-rate hardening is found to be challenging with shell elements, especially w.r.t. dynamic fracture strain, and can increase the uncertainty if not carefully calibrated. Determination of the design material concept is discussed, ensuring that the materials representing load and resistance are given appropriate safety factors. A novel way of separating two different mesh scale effects termed geometric and material is proposed. The micromechanical process of fracture is discussed, and related to the macromechanical process that can be captured with coarse shell elements. The dependence of strain-state and length scale is investigated. Many of the popular fracture criteria are implemented in LS-DYNA and a large simulation program is conducted with different experiments and many mesh sizes. The accuracy of the fracture criteria is assessed. An extension of the BWH fracture criterion is proposed, in which post-necking effects can be included. Through a combination of coupled damage and a strainstate dependent erosion criterion, a more robust fracture prediction with reduced mesh dependence is achieved. NLFEA is used as virtual experiments to study supply vessel collisions with offshore platforms. A pressure-area relation is proposed for the pressures required to initiate crushing of the bulbous bow of a supply vessel. A simplified model is proposed for strength-design of stiffened panels, combining a roof-top mechanism with the stiffener shear capacity in an incremental approach. A refined method for strength-design of jacket legs and braces is proposed, based on the characteristic strength Rc. Various challenges related to simulation of ice collision events are discussed, especially with respect to the material behavior of ice during fast compressive loading. The material behavior of ice is not well known, and has large statistical variations in properties. During a collision with damage to the ship, the changed geometry of the ship side will increase the confinement of the ice, thereby increasing its crushing pressure. All ice loads given in current rules and standards disregard this effect. Two experimental campaigns are performed to study the coupled deformation process as both ice and structure deforms, thereby investigating the effect of the confinement. The load exerted from the ice increases significantly when the structure deforms plastically, indicating that the coupled effect is important. The effect should be considered in updated rules and standards