Don Spencer

Time-Domain VIV Prediction of Marine Risers

Vortex Induced Vibration (VIV) of marine risers poses a significant challenge as the offshore oil and gas industry moves into deep water. A time-domain analysis tool has been developed to predict the VIV of marine risers based on a forcing algorithm and by making full use of the available high Reynolds number experimental data. In the formulation, the hydrodynamic damping is not treated as a special case but simply an extension of the experimentally derived lift curves. The forcing algorithm was integrated into a mooring analysis program based on the global-coordinate based finite element method. At each time step, the added mass, lifting force and drag force coefficients and their corresponding loads are computed for each element. Validation studies have been carried out for a full-scale rigid riser segment and a model-scale flexible riser. The numerical results were compared with experimental data and solutions by other programs.Copyright

Numerical Simulation of Ship Maneuvering in Pack Ice

This paper presents preliminary results from a computer program for simulating ship maneuvering in ice covered waters. The program is derived from two programs, Ship Maneuvering Laboratory (SML) and a discrete element numerical modeling program (DECICE). SML is an in-house code developed by Oceanic Consulting Corporation for simulating ship maneuvering in open water. It is based on a ship maneuvering model originally developed by the Japanese mathematical maneuvering group (MMG). DECICE is a discrete element method which was developed by INTERA Technologies and is used to calculate the ice loads on the ship and the interactions between ice pieces. The paper presents a summary of the mathematical methods used together with the results of some case studies for ships EXM004, PSM004 and Esso Osaka Tanker. These computer predictions include turning circle and Zig-Zag maneuvers. Comparisons and discussion of the simulated results between cases with and without ice are also provided.Copyright

Numerical Vortex-Induced Vibration Prediction of Marine Risers in Time-Domain Based on a Forcing Algorithm

Vortex-induced vibration (VIV) of marine risers poses a significant challenge as the offshore oil and gas industry moves into deep water. A time-domain analysis tool has been developed to predict the VIV of marine risers based on a forcing algorithm and by making full use of the available high Reynolds number experimental data. In the formulation, the hydrodynamic damping is not treated as a special case but simply an extension of the experimentally derived lift curves. The forcing algorithm was integrated into a mooring analysis program based on the global coordinate-based finite element method. At each time step, the added mass, lifting force, and drag force coefficients and their corresponding loads are computed for each element. Validation studies have been carried out for a full-scale rigid riser segment and a model-scale flexible riser. The numerical results were compared with experimental data and solutions by other programs.

Benchmarking of VIV Suppression Systems

A new form of Vortex-Induced Vibration (VIV) suppression device, the AIMS Dual-fin Flow Splitter (ADFS), has been developed, tested and benchmarked against bare-pipe, 5d and 15d pitch strakes and conventional teardrop fairings. Testing included high-mode number in-situ tests as well as low Reynolds number (<300,000) and high Reynolds number (<1.9 million) forced and free tank tests. Finally, wind tunnel tests and in-water Particle Image Velocimetry (PIV) were used to test the hypothesis that the dual-fin flow splitter replaces the oscillating wake of a blunt body with a stable, attached circulation behind the body and between the fins. Such a replacement was hypothesized to result in reduced drag, and the elimination of almost all VIV. The paper will describe the testing program and results, and present the incorporation of the test results into riser models.Copyright

Dynamic Positioning of Vessels Using a UKF-Based Observer and an NMPC-Based Controller

This paper presents a solution to the problem of dynamic positioning (DP) of vessels in ice-infested environments using a nonlinear observer and a finite-horizon optimal control-based approach. An unscented Kalman filter-based nonlinear observer is developed to estimate both the vessel states and the unknown inputs, such as the ice load. To perform better control and disturbance rejection, a nonlinear model predictive controller is employed for DP. The developed modules are integrated with a commercial vessel maneuvering software, and promising real-time results are generated.Note to Practitioners—Accurate dynamic positioning (DP) of vessels in the presence of environmental disturbances is very important for Arctic explorations. This paper proposes a tightly coupled approach to estimate the unknown forces acting on a vessel and the vessel states. Also, an optimum control-based solution is proposed for robust DP.

Transient Hydrodynamic Forces on a Disconnectable Turret Buoy

For disconnectable turret-moored FPSOs, accurate prediction of turret buoy and FPSO motions during the buoy disconnection process is essential for safe operations. For deepwater high production rate systems, large size buoys are required to accommodate the large number of risers and heavy mooring legs. Analytical models of hydrodynamic forces on large size buoys must be verified before they are applied to motion predictions. To gain a better understanding of the transient hydrodynamic loads on the buoy and hydrodynamic interactions between the buoy and the hull during disconnection, we conducted a specially designed model test in a tow tank. In the model tests, both the buoy and the FPSO models were forced to oscillate by two independent actuators in calm water and in waves. Summary of test results, computed transient hydrodynamic forces from a simplified approach, a true time-domain transient hydrodynamic analysis based on instantaneous buoy positions, and computational fluid dynamics (CFD) results are presented in this paper.Copyright

Loads due to First Year Ice Ridges on a Vertical Cylinder

This paper presents some predictions of ice ridge loads on a vertical cylinder, based on model experiments, analytical methods, and numerical simulations. The experiment program involved the interaction of a vertical cylinder with two unconsolidated ice ridges. Each ridge was nominally 1.9 m wide with a keel 0.5 m deep. The nominal scale used for the experiment was 1:30. The analytical methods included Dolgopolov, Mellor, Croasdale & Cammaert, and ISO19906. The numerical simulations were carried out using a discrete element computer code, named DECICE. The paper compared the results of model experiments, analytical methods and numerical simulations.Copyright

VIV Response of a Subsea Jumper in Uniform Current

Subsea jumpers are susceptible to in-line and/or cross-flow vortex induced vibration (VIV) fatigue damage due to sea bottom currents. However, there is no proven industry standard design analysis methodology currently available specifically for assessing subsea jumper VIV response.In 2012, ExxonMobil conducted a jumper VIV model test to assess the validity of potential jumper VIV prediction approaches. A towing test rig was used to expose a small scale jumper model to flow conditions simulating uniform bottom currents. The jumper model was instrumented to acquire acceleration, bending strain and end connection load data. Several accelerometers and strain gauges were installed to enable reconstruction of static and dynamic deformations and bending deflections along the jumper model. Towing tests at different orientations and tow speeds were performed on both a bare pipe model and a straked pipe model. The data were analyzed to examine the frequencies and amplitudes of the jumper vibration. The data from these experiments provide a benchmark for validating jumper VIV prediction approaches.In this paper, the model test program is presented including model testing philosophy, jumper design and fabrication, and high level model test results.Copyright

Tandem Riser Hydrodynamic Tests at Prototype Reynolds Number

Deepwater riser interference is an area of significant technical complexity and uncertainty in the design cycle due to the intricacies of wake hydrodynamics. Existing models, found in industry guidelines, are based on approximate theoretical models of bare cylinder wake and nominally checked against small scale tests at low Reynolds numbers. In actual conditions the Reynolds number is sufficiently higher and the risers are fitted with vortex-induced vibration (VIV) suppression devices. This raises questions on the applicability of the standard models and hydrodynamic coefficients used, especially if the geometry is different than a circular cylinder. A series of full scale tests, at supercritical Reynolds numbers, were conducted to address these uncertainties and obtain hydrodynamic coefficients for interference design. The tests were carried out utilizing two full scale cylinders fitted with actual VIV suppression devices and towed either in fixed or spring supported configurations. The paper discusses the experimental methodology and findings from the testing program, showing deviations from the standard models found in industry codes.Copyright

Sheared Current Generation in Flume Tank for Experimental Research

Current velocity, profile, direction, and duration may affect hydrodynamic loads and VIM of offshore structure. It is often recommended that physical experiments are carried out in sheared current, in multiple directions and for sufficiently long period of time to investigate the hydrodynamic characteristics of deep draft offshore structures to obtain better correlation to the field measurements. This necessitates generating sheared current with acceptable turbulence level. This paper presents a recent advancement in generating sheared current in a flume tank facility. In this process, the test specimen remains moored and the water flows past with its velocity varied with depth as long as necessary.A combination of synthetic and wire meshes are used to provide the required amount of blockage onto the circulating channel flow of the flume tank to obtain specified current distribution across the cross-section and at the longitudinal center of the tank. The final set-up of the current screen provided a sheared flow distribution within 10% of the targets. Also, the measured turbulence level was below 10% in all the locations measured.VIM studies of a model spar were successfully carried out in the generated sheared current in the flume tank facility. The ability to accurately model the sheared flow essentially improves the accuracy of the measured VIM type response measurements. The generated sheared current can also be applied for other hydrodynamic experiments where sheared current is relevant.Copyright