Abstract
Underwater turbines are an interesting solution for future energy demands. An important design consideration is the potential for impact with their surrounding natural environment. This work focuses on the dynamic event of impact with a large sea creature, the minke whale is chosen as a test case as it is a well-established species in the waters being considered. The simulation goals are to capture the macroscopic event of whale to turbine impact whilst both are submerged in fluid, with the primary focus being the hydrodynamic effects of the fluid on the whale. A particle based elastic modelling technique named lattice spring modelling (LSM) that is well suited to implementation on a graphics processing unit (GPU) is presented, this is coupled with smoothed particle hydrodynamics (SPH) to create the SPH-LSM method. The proposed solution offers unique capabilities when compared to alternatives. This approach dynamically assigns identity to particles depending upon their current position, with an interface formed by particles that move according to different sets of governing equations. The LSM presented aims to capture a linearly elastic model using first order Hookean response, however the basic methodology can be used as a framework to support more complete constitutive models. The SPH and LSM methods are described and their GPU implementation detailed, including a method to derive a spring constant based on a Young's modulus and Poisson's ratio. A test case is provided in the form of a 3-D beam deflection test and results compared favourably against Timoshenko beam theory. A preliminary case showing an impact event between a geometrically realistic minke whale described using LSM and a rigid rotating turbine is presented, showing the method to be technically viable.