Numerical modeling of nonlinear Newwaves for impact assessment on offshore structures

Numerical modeling of nonlinear Newwaves for impact assessment on offshore structures

Burger, W.J.

Stelling, G.S. (mentor)
Pinkster, J.A. (mentor)
Reniers, A.J.H.M. (mentor)
Ewans, K. (mentor)
Sliggers, P.G.F. (mentor)
Forristall, G.Z. (mentor)
Loots, G.E. (mentor)

Civil Engineering and Geosciences

Hydraulic Engineering

2005-01-01

Offshore structures are designed to resist the impact of the extreme wave. Shell EP Projects desires a computer model that is able to simulate the impact of the design wave on gravity based offshore platforms, such as the Sakhalin PA-B platform. Comflow is a numerical model that simulates fluid flow, by solving the Navier-Stokes equations using the improved Volume-of-Fluid method. The wave inflow is prescribed at a side boundary of the Comflow domain. Shell uses the Newwave theory to prescribe the wave group that contains the design wave. The shape of the wave train that contains the design wave is important for the impact on the structure as it influences wave run up and reflection; possibly causing the wave to slam against the bottom of an offshore platform deck. Shallow water and wave steepness influence the shape of the wave group; this can be simulated with nonlinear wave models. This thesis concerns the simulation of nonlinear Newwaves and the implementation in Comflow. The goals of the thesis hence are to: 1. Find a wave model to simulate fully nonlinear Newwaves. 2. Couple the nonlinear wave model to Comflow. 3. Validate the wave impact on structures in Comflow with scale model tests. Vijfvinkel [25] developed a quasi-spectral method to solve the wave potential equation; the one dimensional spatial domain is periodic and it requires solely an initial condition. The method is based on expansion of the Dirichlet-Neumann operator, which is substituted in the governing equations. The operator consists of the surface elevation and the velocity potential at the surface. These are the properties that are calculated in a time domain. A velocity profile can be reconstructed. The Vijfvinkel computer code was obtained and revitalised. The Beji Battjes test [3] was simulated in Comflow. It confirmed the grid and time step requirements [19], and it showed that the nonlinear formation and release of higher harmonics are also simulated correctly following the requirements. Three additional numerical tests were set up to indicate the validity of Newwave simulation. Each test highlights a separate aspects of nonlinear modeling of Newwaves. • The second experiment consists of a tank in which a standing wave is sloshing. The simulation in Comflow is done for various relative water depths. It showed that, except for shallow water, wave propagation is simulated reasonably well, however, damping can play a significant role due to the upwind discretisation method and velocity extrapolation method at the surface. • In the third experiment the wave steepness is increased up to near breaking conditions, in Comflow and in Vijfvinkel. It proved that Vijfvinkel has superiour performance in simulation of steep waves. • In the fourth experiment Newwaves are simulated in Vijfvinkel. A method is used that transforms the linear initial condition to a quasi-nonlinear initial condition. Simulations with a starting time of more than 50 seconds prior to the linear focus time, have higher peaks before the wave group reaches the linear focus location. The wave group becomes a bound wave group; the shape of the wave group shows less variation, and it propagates with a common velocity. Vijfvinkel is coupled to Comflow. The output of Vijfvinkel is transformed to comply with the Comflow domain, and the input procedure of Comflow is extended to be able to read external input files. This new input procedure allows the simulation of an arbitrary disturbance of the free surface. A simulation is performed in which Vijfvinkel prescribes the nonlinear Newwave to the inflow boundary in Comflow. A necessity is the development on the numerical core of Comflow. This was identified and anticipated on in the Joint Industry Project, as a large number of grid cells is required to model propagating waves and damping plays a significant role. For the simulation of impact on offshore structures the implementation of local grid refinement will increase the accuracy and efficiency of the simulation of wave impact on structures. Only when these developments are achieved, validation with scale model tests is sensible, for Comflow to become a valuable design tool to assess wave impact on offshore structures.

Comflow
Sloshing
standing waves
nonlinear newwave
Vijfvinkel

http://resolver.tudelft.nl/uuid:dbea96ed-4554-48bc-ad7e-51b22742c41b

Student theses

master thesis

(c) 2005 Burger, W.J.