The vibration assessment of subsea systems has largely been limited to vortex-induced vibration of riser systems and unsupported pipeline spans (i.e., environmental loading) caused by flow past the outside of a riser or conductor. But now, piping vibration caused by process excitation is becoming an issue on manifolds and jumpers, in part associated with increasing production rates. Though the likelihood of a failure is low, the consequences can be high, resulting in an unacceptable level of risk. As well as piping integrity issues, additional vibration-related problems have also been experienced with valves and instrumentation.
Process Excitation Mechanisms
There are a number of different sources of excitation and these generally depend on the type of process fluid within the system. The following three mechanisms are the most common.
Flow-Induced Turbulence. This is caused by broad band, low-frequency energy, generated by “single phase” turbulent flow through valves, expansions and bends—essentially anything that disturbs the flow. This can lead to excitation of the low-frequency modes of the piping system by energy transfer from fluid momentum to the pipe, resulting in low-frequency vibration (up to possibly 50 Hz). This is widespread in most processes and depends heavily on the velocity and density of the fluid.
Multiphase Flow. This encompasses slug, bubble, annular, and churn flow. The frequency range of the resulting excitation is heavily dependent on void fraction and flow regime.
Flow-Induced Pulsation. This is linked to gas flow through a flexible riser or jumper, sometimes known as the “singing riser.” This is where gas across the internal corrugated geometry within the flexible piping causes vortex generation. This can cause a pressure fluctuation, or pulsation, with a frequency and amplitude that are dependent on the gas velocity. This pulsation can drive the flexible piping at either end at very high frequencies—in some situations as high as 1,000 Hz—causing fatigue damage to accrue very quickly. The phenomenon is typically only experienced on dry gas systems, so it can be an issue on gas export, gas injection, and gas lift systems.
The Hidden Threat
Pipe design and construction can present a number of unique challenges in subsea developments, given their remoteness and complexity to access. A vibration issue may occur subsea without any obvious sign topsides.
One of the key challenges is the practical difficulty of obtaining subsea measurements on equipment installed on the seabed. This can be very expensive, usually requiring a dive support vessel to be on station to deploy and recover the monitoring instruments. For fatigue assessment, direct strain measurement usually would be preferred, but it is often impossible to perform on subsea equipment that has already been deployed, such as production manifolds, pipeline end manifolds, and flowline termination assemblies. In practice, we often must measure the vibration acceleration at nonideal locations and infer the system fatigue damage by linking the measurement data with a suitable simulation.
Another challenge relates to the often limited real-time capability of the measurement instruments. Fatigue damage can accrue relatively quickly. The delay between taking the measurement and recovering and analyzing the vibration data—which may be days or weeks—often results in an unacceptable risk.
New Industry Guidance
To counter some of these problems, Xodus Group has acted as a technical author of a soon-to-be-published Energy Institute (EI) document that builds on the topsides version of EI’s guidelines for piping vibration—providing a risk-based methodology and good design practice—and tailors an approach specifically designed for subsea equipment. The anticipated document included input from a wide variety of operators, equipment designers, and consultants.
This latest EI guidance, which is slated to be released this summer, is based solely on subsea equipment. The document includes a step-by-step approach to identify and solve potential issues early in the design cycle to achieve a fatigue-resistant design. There are a number of key elements.
Defining Good Design Practice. This includes guidance on the design and layout of process piping, how branch connections should be made, and advice on rough-bore flexibles.
Initial Risk Assessment. This includes screening methods using information from piping and instrumentation diagrams, process flow diagrams, and in some cases, piping isometrics. The existence of potential hot spots stemming from any of the common excitation mechanisms is identified on the build being assessed, considering the static and dynamic response of the piping. The method uses a series of algorithms and process and piping information to ascertain a likelihood of failure (LOF) score for each excitation mechanism. Responsive actions are determined based on the LOF score. The method can be used to check the sensitivity to changes in the process conditions and/or gross structural parameters (e.g., pipe diameter, wall thickness, and degree of support).
Detailed Simulation. If the fatigue risk is not acceptable, guidance is provided on how to perform a detailed fatigue simulation for a variety of excitation mechanisms. Once the level of excitation is predicted, using empirical, analytical, or computational fluid dynamics (CFD) techniques, a finite element model is created to assess the response of the piping to the vibration acceleration and dynamic stress. This requires knowledge of the piping system layout, supports, process conditions, and structural boundary conditions (Fig. 1).
The results provide an assessment of the fatigue life under certain operating conditions and the effect of varying those conditions. In addition, the results help to identify the best locations to place vibration sensors for subsequent monitoring activity. However, the industry still needs better validation data to ensure accurate simulation methods, particularly for excitation generated by multiphase flow.
Monitoring System Specification. There are some key considerations when specifying the monitoring system. They include the frequency and dynamic range, as well as the phase relationship between sensor pods and how transducers will be installed and mounted to parts of the structure. Most important is how the data will be recovered and used.
Verification. Guidance is provided on the verification steps required at the construction stage to ensure that what is built correlates with the intended design. This is particularly important for piping supports, which have an important bearing on the vibrational response of the system. Verification may also include nonintrusive modal testing of the piping systems and associated support structures. This will help to provide validation data for any previous simulations. If used with temporarily installed strain gauges, this type of testing in the construction yard can also provide useful information about the relationship between the vibration acceleration levels at locations where measurements will be made on subsea infrastructure, and the dynamic stresses at potential fatigue hot spots that could not be monitored subsea because of access restraints.
Monitoring During Operation. This includes checking of the raw data (signal statistics, including kurtosis), data interpretation, and the interface with simulation to provide an assessment of the fatigue performance of the equipment under different operating conditions. An example of a fatigue assessment based on the data logged from subsea vibration sensors is shown in Fig. 2.
Looking Ahead
As flow rates increase, the industry must better understand the types of dynamic forces that are generated by multiphase flow in piping systems. A new joint industry project (JIP) managed by Xodus addresses this area of uncertainty, with support from key industry operators and equipment vendors.
This initiative will derive, from a series of “industrial scale” tests, the power spectral density (PSD) of the forces acting on piping under a variety of void fractions and superficial velocities, designed to include the most common flow regimes. This would act as validation data for CFD predictions and could provide an empirical means of determining the force PSD acting on a pipe bend, given basic flow parameters. It is also intended that this new JIP will provide a series of benchmarks for the use of CFD to predict two-phase flow forcing functions.
Conclusion
Following the Macondo incident in 2010, the integrity of subsea systems is receiving increased attention. Research into vibration-induced fatigue and its management in the subsea environment is attracting greater emphasis and commitment in the industry.
As flow rates increase and exploration and production go into deeper, harsher environments, subsea equipment is becoming more complex to enable improved production performance. New developments in subsea processing and separation are also on the horizon. Finding the balance between design, simulation, and monitoring is crucial to maintaining the integrity of equipment and to pushing ahead to recover more reserves.
This article is taken from a paper presented at the 2013 Australasian Oil & Gas Subsea Exhibition and Conference, 20–22 February in Perth, Australia.