Numerical Simulation Of Wave Run-up And Slamming On Multi-Pod Offshore Wind Turbines In Nonlinear Regular Waves And Random Waves

The modern development of computer hardware and software technologies have seen the soaring CPU computing capacity, wide application of multi-core parallel computing, larger size and higher speed of memory and hard disks, and faster numerical algorithms. All these have rendered the computational fluid dynamics (CFD) with unique advantages to aid the theoretical research and experimental study. It is the objective of this study to employ CFD as an efficient tool to numerically investigate the wave run-up and slamming mechanisms on monopile, tripod and quinque-pod foundations of offshore wind turbines. Different from the commonly used Airy wave, this study focuses on nonlinear regular waves and random waves.The height of wave run-up not only determines the air-gap, but also is associated with the magnitude of instant slamming loads on working platform and the stairs on foundation. Due to the complex fluid mechanisms of run-up and slamming, this pioneering study turns to the CFD tool, while the comparison with model tests will be also carried out for benchmark purpose.First, using CFD software FLUENT and laminar flow, simulation of linear waves is achieved in the numerical wave flume by defining the far-field inlet boundary conditions and by VOF solver. Comparisons with wave profiles in physical wave flume show that the simulated waves are accurate. The meshing strategy is also discussed. It follows that the simulation of fifth-order Stokes waves and linear random waves using JONSWAP wave spectrum and FFT recovery are successfully realized in the numerical wave flume, with reasonable computing efforts and good accuracy. To categorize the model for a prescribed wave condition, the Ursell=40 criterion in Fenton (1998) is recommended for use. Zones to the left of this criterion are for Stokes waves while the right zones are for cnoidal waves. For wave conditions nearby the border, the Stream Function is suggested for calculating the far-filed wave kinematics.Second, simulation are carried out for the wave run-up on monopile, tripod and quinque-pod foundations under a variety of nonlinear regular wave and random wave conditions. For each type of foundation, two water depths are considered. Comparison of CFD results against the model test data has shown good agreement.The buildup progress of run-up on the tripod foundation in a relative deep water with a nonlinear regular wave are demonstrated for the complex fluid mechanism. Also, for the quinque-pod founfdation in a relative shallow water, the buildup progress of run-up is illustrated for the difference between multi-pod and monopile foundations. An important observation from the CFD simulations is that the lowest run-up occurs at the radical direction of 145°, rather than 1350 reported in literature. This is useful for fixing the direction of stairs attached to the foundation in the early design as well as for offshore landing in directional waves. For random waves, the wave run-up coefficient m is obtained by using the method of equivalent largest wave in the time series, though this method needs to be validated in future work. The regression analysis of m shows that its trend in multi-pod foundations differs from the traditional monopile foundation. Even the trend in tripod foundation is different from that in quinque-pod foundation. Factors like water depth and wave slope matter.Finally, slamming on the landing platforms of tripod and quinque-pod foundationsis simulated in the established 3D numerical wave flume for nonlinear regular waves and random waves. The CFD method is again proven to be a feasible approach for simulating a complicated fluid phenomenon. However, The slamming coefficient Cs in simulations tends to be smaller than that in model tests. The difference is greater than expected and further-study using turbulent models might need to be employed in CFD simulations.Coefficients m and Cs are interested numbers for the engineering practice. The regression analysis of them in this study through CFD simulations has shown that the Ursell parameter is more useful than wave slope in constructing the fitting curves for the trend of m, which provides another meaningful reference for the design and safety assessment of offshore wind turbines.