| dc.description.abstract | Bridges are a necessity for transportation across waterways in modern society, and the passing of large ships below bridges poses the risk of ship-bridge collision, which can, in the worst case, lead to the total collapse of the bridge with fatalities and high economic costs. Ship-bridge collisions have been studied extensively over the past years to understand and reduce these risks.
This thesis study how different impact conditions related to the impact velocity, ship mass, and impact angle affect the collision and the probability of collapse. It includes numerous simulations using different impact conditions and proposes models that relate the impact conditions and collision response to the likelihood of collapse, modelled as the Heinrich Factor.
The first step was to get to familiar with the finite element software LS-Dyna through an introductory course and self-studies. The finite element model of the bridge structure was created in LS-Dyna, and ship-bridge collision simulations with different impact conditions were performed. The simulation results were analysed and used to determine the probability of collapse of the bridge pier. Additionally, two scripts in Python were developed to apply machine learning principles to predict the collapse of a bridge pier, using the simulation results as data. | |
