CFD Simulation of a Riser VIV
by
Hamn-Ching Chen and Kevin Huang, Ocean Engineering Program, Department of Civil Engineering, Texas A&M University
Chia-Rong Chen, Department of Mathematics, Texas A&M University
Richard S. Mercier, Offshore Technology Research Center

December 2007

 

Vortex-induced vibration (VIV) is an important issue in the design of deepwater riser systems, including drilling, production and export risers. The VIV can produce a high level of fatigue damage in a relatively short period of time for risers exposed to severe current environments. The wake interference between various risers in the same riser array may also lead to collisions between adjacent risers. As some of the recently discovered reservoirs are located in a water depth near 10,000 ft (3,000 m), it is desirable to develop advanced computational fluid dynamics (CFD) tools that can provide reliable prediction of riser VIV in ultra deep water environments.

Partially driven by the need for offshore oil and gas production in deepwater fields, numerical simulation of riser VIV has been an active research area in recent years. Experiments are sometimes preferable to provide design data and verification. However, deepwater risers have aspect ratios that are so large that model testing is constrained by many factors, such as experimental facility availability and capacity limits, model scale limit, difficulty of current profile generation, and cost and schedule concerns. Under such conditions, CFD simulation provides an attractive alternative to model tests. The advantages of CFD simulation are obvious:

1. It is less sensitive to the riser length, therefore the water depth is not a technical barrier for the CFD approach.
2. Every aspect of the riser VIV phenomenon can be analyzed, including riser global motion behavior, vortex shedding details, drag and lift force components, etc.
3. Complex flow fields (due to the existence of nearby risers or hull structures, for example) and current profiles (such as submerged or bottom currents) can be readily handled.

Note that the deepwater current profiles tend to be more complex than in shallow water. For example, the typical loop current eddies in the Gulf of Mexico are usually submerged several hundred meters underneath the surface, while in some fields in offshore West Africa and offshore Brazil, the current may reverse direction along the water column. Obviously not all of these current profiles can be easily simulated in a wave basin. The CFD approach provides a cost effective alternative to evaluate the riser VIV and related issues under these complex current conditions. With a validated CFD code the complexity of the current profiles can be readily accommodated, usually through changing the far field fluid velocities and boundary conditions. Nevertheless, the disadvantage of the CFD approach is also obvious – it is very time consuming, even with the help of the fastest computers and parallel computational technology.

Many software tools have been developed in the oil and gas industry to perform riser VIV analysis. However, the majority of them are based on empirical formulas, heavily relying on model test data. This approach could provide satisfactory VIV predictions for shallow water risers, where their length over diameter ratio (L/D) is fairly small, and model tests could be easily carried out to provide input data and/or verification. Deepwater risers are likely to have high order mode vibration in strong currents. Under such conditions, model testing in a wave tank is difficult due to tank size or model scale limitations, while field experiments are feasible but costly. Furthermore, there are some important characteristics associated with deepwater riser VIV yet to be studied and understood, such as:

1. deepwater risers tend to experience multi-mode vibration, therefore it would be overly conservative to assume single-mode lock-in, and
2. the excited modes in deepwater riser VIV could be very high, while higher modes are more sensitive to damping, hence showing strong nonlinear behavior.

In a word, time domain CFD simulations are very promising and appropriate for deepwater riser VIV analysis.
There are numerous experimental and numerical investigations on the subject of a circular cylinder undergoing vortex-induced vibrations (VIV). Blevins (1990) summarized some of the early research work on flow induced vibrations. Govardhan and Williamson (2000) reviewed some experimental assessment of vortex formation modes. Some of the VIV studies on low mass ratio cylinders have been reviewed by Willden and Graham (2004). Various VIV numerical investigations have been reviewed by Dong and Karniadakis (2005). Lucor et al. (2006) reviewed some research work on complex modes. Some existing CFD codes for riser VIV analysis have also been reviewed and compared in Chaplin et al. (2005). Trim et al. (2005) presented experimental details for a long riser under various current conditions. Holmes et al. (2006) used a fully 3D simulation approach to analyze riser VIV and the effect of strakes. Several other existing CFD codes for practical riser VIV analysis were reviewed by Chaplin et al. (2005).

Over the past several years, we have developed a Finite-Analytic Navier-Stokes (FANS) computer code for riser VIV simulations (Chen et al. 2006) at Texas A&M University. Some of the previous applications of this code include:
• 2-D simulations of flow past a fixed riser at high Reynolds numbers,
• surface roughness effects,
• 2-D simulations of elastically mounted risers undergoing VIV at high Reynolds numbers: single isolated riser and arrangements of multiple risers,
• 3-D large eddy simulation of flow past a fixed riser,
• 3-D large eddy simulation of an elastically mounted riser undergoing VIV,
• simulations of an elastically mounted riser outfitted with a fairing, and
• simulations of an elastically mounted riser with helical strakes.
The above simulation results clearly demonstrated the capability of the FANS code for time-domain simulation of VIV responses of 2D and short 3D (L/D ~ 10) risers at high Reynolds number with or without VIV suppression devices. In this report, the FANS code has been further extended for 3D simulations of long and flexible marine risers with L/D up to 3,000.

It is well known that the riser VIV responses are affected by many parameters including the Reynolds number, surface roughness, strakes, fairings, 3D sheared currents and ambient turbulence. In order to provide accurate analyses of the VIV phenomena, the Finite-Analytic Navier-Stokes (FANS) numerical method has been employed in conjunction with a chimera domain decomposition approach to investigate the complex deepwater riser VIV induced by various current profiles. As noted earlier, the FANS method has been successfully used for VIV analysis of smooth and roughened risers in uniform currents. In this research, the method has been further extended for the prediction of VIV responses of deepwater risers under both the uniform and sheared current profiles. The simulation results were compared with available experimental data to assess the accuracy of the CFD predictions.

In order to extend the predictive capability of the FANS code from relatively short 3D risers with L/D ~ 10 to long 3D risers with L/D ~ 1,000, the following numerical investigations have been performed and summarized in this report:
• development of modal solver for riser finite element motion equation,
• development of direct solver for riser finite element motion equation,
• 2-D simulations of flow past a fixed riser at high Reynolds numbers,
• 2-D simulations of flow past a forced motion riser at high Reynolds numbers,
• 3-D simulations of flow past a horizontally positioned riser in uniform current,
• 3-D simulations of flow past a horizontally positioned riser in shear current,
• 3-D simulations of flow past a vertically positioned riser in uniform current,
• 3-D simulations of flow past a vertically positioned riser in shear current,
• validation of FANS simulation results with experimental data, and
• comparison of FANS results with numerical results obtained by commercial codes.

The simulation results clearly demonstrate the capability of the FANS code for accurate prediction of VIV responses of deepwater risers under uniform and sheared currents.