Fully three dimensional computational fluid dynamics (CFD) solutions are combined with structural models of a tensioned riser to predict riser vortex induced motion. The use of three dimensional CFD solutions overcomes many of the shortcomings of combining a series of strip or two dimensional simulations to calculate the fluid forces on the riser. Three dimensional vortex structures are treated correctly and straked risers and variations in angle of attack can be studied directly. The proposed method uses finite element methods that are tolerant of sparse meshes and high element aspect ratios. This allows economical solutions of large fluid domains while retaining the important features of the large fluid vortex structures which drive risers. Long risers can be treated with readily available computers and examples of simulations of riser with L/D over 1400 are given and compared with previously published experimental data. These examples are used to illustrate several points regarding the effects of the treatment of the riser structure as well as the efficacy of rotating frame or pinned riser experiments used to simulate sheared currents. The method can also be extended to sheared currents whose heading varies with depth.

Floating spar platforms are widely used in the Gulf of Mexico for oil production. The spar is a bluff, vertical cylinder which is subject to Vortex Induced Motions (VIM) when current velocities exceed a few knots. All spars to date have been constructed with helical strakes to mitigate VIM in order to reduce the loads on the risers and moorings. Model tests have indicated that the effectiveness of these strakes is influenced greatly by details of their design, by appurtenances placed on the outside of the hull and by current direction. At this time there is limited full scale data to validate the model test results and little understanding of the mechanisms at work in strake performance. The authors have been investigating the use of CFD as a means for predicting full scale VIM performance and for facilitating the design of spars for reduced VIM. This paper reports on the results of a study to benchmark the CFD results for a truss spar with a set of model experiments carried out in a towing tank. The focus is on the effect of current direction, reduced velocity and strake pitch on the VIM response. The tests were carried out on a 1:40 scale model of an actual truss spar design, and all computations were carried out at model scale. Future study will consider the effect of external appurtenances on the hull and scale-up to full scale Reynolds’ numbers on the results.

Fully three dimensional fluid flow simulations are used with a simple structural model to simulate very long risers. This method overcomes many shortcomings of methodologies based on two dimensional flow simulations and can correctly include the effects of three dimensional structures such as strakes, buoyancy modules and catenary riser shapes. The method is benchmarked against laboratory and offshore experiments with model risers of length to diameter ratios up to 4,000. RMS values of vortex induced vibration motions are shown to be in good agreement with measurements. The resources needed to model ultra deep water drilling and production risers are estimated based on current computer technology.

Submersibles used for exploration, maintenance and naval warfare have to be both manoeuvrable and easy to control. Simulation of the trajectory for these vessels requires the accurate determination of the hydrodynamic forces and moments which are determined by model-testing, empirical methods or a combination of both. CFD can play a role here by permitting an easier and more accurate determination of these loads. In this paper we focus on the accurate prediction of the manoeuvring forces of free swimming streamlined submersibles (submarines) using CFD. We compare simulations of a standardised well-known submarine shape (DARPA SUBOFF) for two configurations, one bare hull (AFF-1) and one fully-appended hull (AFF-8), under different inflow angles. The viscous-flow solvers used are the finite volume solver ReFRESCO developed by MARIN, and the finite element commercial solver AcuSolve. Verification studies are performed and the numerical results are validated with the experimental data available in the literature. The influence of different turbulence models is investigated and results obtained with a RANS (Reynolds-Averaged-Navier-Stokes) approach are compared with the theoretically more realistic DDES (Delayed-Detached-Eddy-Simulation) results. The influence of the appendages on the forces and flow fields is also investigated and analysed. As a last example, results of a forced pitch motion including dynamic effects are presented.

This paper describes measurement of the aerodynamic pressures produced by a Bombardier High-Speed Non-Electric Locomotive (HSNEL) on an adjacent stationary double-stack freight car. Static pressures are measured on the near and far-side faces of the freight containers over a range of locomotive speeds from 60 to 130 mph. This data is also compared with the pressures predicted by computational fluid dynamics (CFD) simulations.