Finite Element Analysis of Ship Collisions
MetadataShow full item record
The serious consequence of ship collisions necessitates the development of regulations and requirements for the subdivision and structural design of ships so that damage and environmental pollution is reduced, and safety is improved. A simplified collision model (SIMCOL) is currently being developed at Virginia Tech to conduct probabilistic analysis of damage in ship collisions and ultimately optimize ship structural designs to improve crashworthiness. Collision data for validation of SIMCOL is very difficult to obtain, and model testing is very costly. Finite Element Analysis (FEA) provides an alternative to physical validation that can be used to increase confidence and insight in simplified model results. This thesis develops a complete methodology for ship-to-ship collision simulations using the explicit non-linear FE code LS-DYNA3D. Various modeling alternatives are considered. The ability to model a complete ship-to-ship collision is developed incrementally starting with bow collisions with a rigid wall. A super-element (intersection model) approach is considered to increase the calculation speed of bow models. A conventional fine mesh finite element bow model is also developed. Results from both models are compared with each other, and with a closed-form calculation from Pedersen. A fine mesh model is developed for a struck tanker cargo section and integrated in a total ship framework modeling external dynamics and ship-to-ship contact. A series of collision scenarios is simulated using the conventional bow model and a rigid bow model striking a double hull tanker. Results are compared with SIMCOL, DAMAGE, DTU and ALPS/SCOL models. LS-DYNA provides detailed and reasonable results for ship collision analysis and comparison to increase confidence in simplified models.
- Masters Theses