Energy Policy Act of 2005 (Ultra-deepwater and Unconventional Resources Program)
Design Investigation of Extreme High Pressure, High Temperature (XHPHT), Subsurface Safety Valves (SSSV) Graduate Student Design Project
This project will investigate subsurface safety valve (SSSV) design in support of the development of new designs for application in valves subject to pressures of 30,000 psi and temperatures in excess of 400° F. Such extreme high pressure, high temperature (XHPHT) conditions are expected in ultra-deepwater areas of the Gulf of Mexico.
Rice University, Houston, TX 77251
Accessing the ultra-deepwater reserves of oil and gas in the Gulf of Mexico (GOM) will require a shift in thinking in regards to the design of sub-surface drilling and production components. This project will examine the design of SSSVs. The extreme high pressure and high temperature conditions that future subsurface safety valves will be expected to perform in require an analysis of current design as a stage in the development of the next generation of SSSVs.
This project is a two-year graduate student design task focused on a flapper-type SSSV. It will utilize a parametric modeling system involving several state-of-the-art software tools. Parametric model creation will allow much of the complexity of the analysis to be quickly applied to a variety of valve design or design modifications.
The design of a flapper SSSV is basically an optimization task constrained by very tight spatial limitations. A series of computer models will hopefully lead to a new design for the XHPHT environment. The structural finite element model will include concepts for: the valve body and connections, flapper & seat geometries, hinge pin and springs, hinge pocket, and flow tube.
The project will examine public information from existing valve manufacturer’s SSSV design(s) and supporting engineering data and will attempt to identify failure modes, pertinent variables for model construction and the impact each variable has on the valve design. Through research of current designs and construction of models using computer software, this project will seek to lay the groundwork for development of an XHPHT SSSV.
The design studies will be documented in detailed annual written reports. Condensed versions of the final design results will be prepared and submitted as an SPE paper. Electronic copies of the data inputs and outputs generated using a variety of commercial software tools will be posted on a supporting web site. New extensions of software developed during this study they will also be posted in source form along with the supporting data files.
The product of this project will be knowledge that can be used to accelerate the development of an improved flapper SSSV design that will in turn reduce the risk of valve failure. Reduced failure risk will result in fewer valve failures after installation and a resulting decrease in shut in production, and valve replacement and maintenance costs.
The research team built and executed a “reservoir to surface” flow model that was then used to evaluate flow at various reservoir depths. The model was used to establish the general flow conditions before any valve closure.
Previously developed solid models of the valve geometry were further refined. The “solid” model of the interior flow domains were extracted and fed into the computational fluid dynamics (CFD) solver. Initial meshes of the flow domains were established.
The initial solid models of the hemiwedge and flapper geometric models involved in the fluid interface we tested for the range of motion from fully open to fully closed. Boolean subtractions were used to then build the complementary interior fluid shapes for each partially closed position. Each of the eight (each) hemiwedge and flapper closure models were meshed in Ansys CFX.
A study of erosion flow theories was completed. It was determined that CFX provided reasonable erosion models, so they will be added to the closure flow studies.
The low cycle fatigue studies have been completed. The stresses resulting from static reservoir pressures will be employed as the cyclic low stress value. The maximum stresses induced in the valve components from various CFD closure positions will be the cyclic high stress value. Since that value depends on fluid properties, a lack of clearly defined XHPHT fluids data is slowing all aspects of the designs.
The draft design reports for the first stage studies for the flapper and hemi-wedge were started. Once completed, the reports and supporting computer files will be moved to a web site for public access.
The key tasks to be undertaken during this project are outlined below.
1. Obtain Public Data on Current SSSV Designs - The design to be developed will have to meet the needs of the major oil producers, such as BP and Shell. These organizations will be the ones that in fact will set the tightest constraints on an optimal design (e.g., the diameter of the tubing most likely to be used in producing an XHPHT ultra-deepwater field). The researchers will meet with operating companies to gain a better understanding of these constraints. Discussions will be held with engineers at Baker Tools and similar SSSV supply firms to gather all the available public data on SSSV products, as well as current design weaknesses. Private discussions have revealed that there currently are problems with SSSV hydraulic controls that can cause them to fail to activate due to high annulus pressure exterior to the valve itself. More information on those potential or actual failures will help in establishing improved designs.
2. Conduct Dynamics Study of Closure Impact Forces - Before undertaking a non-linear computational fluids-structure interaction study the initial flapper dimensions need to be selected. The major flapper stresses come from the impact with the valve seat at the instant of closure. The values of the angular velocity and angular acceleration at that instant will be determined, for current pressure load estimates, by applying classical rigid body analytic mechanics. That will be the first of at least three approaches to establish those data.
The second approach will follow the completion of the project’s solid model of a SSSV assembly. That solid model will be constructed with a SolidWorks modeling system utilizing the first size estimates from the analytic mechanics study. That model will be transferred to a CosmosWorks finite element system. CosmosWorks will load the initial flapper model with the angular velocity and acceleration from analytic mechanics, and the currently assumed pressure loads at the instant of value closure. The results of those studies will be a full, three-dimensional set of stresses and displacements for the flapper, its hinge, valve seat, and surrounding components included in the model. Those data will provide a basis for investigating the possible shapes for the flapper and associated members.
A similar kinetics study will be conducted by transferring the SolidWorks data to the companion CosmosMotion software. It numerically solves the analytic equations of motion, for an elastic body, and yields the time history of the forces and moments acting on the slider, hinges, and springs. Those details are usually more accurate than the analytic estimates, and the results can be automatically imported into CosmosWorks.
The third approach will take the best flapper shape and employ it in a complicated fluid-structure interaction study. That will be accomplished by exporting the SolidWorks model to the Ansys finite element software and its associated CFX computation fluid dynamics solver. Then the actual closing of the valve will be modeled to find an improved set of values for the angular velocity, angular acceleration and non-uniform pressure distributions on the flapper at and after closure. Those results are influenced by the developed flapper shape, and are expected to show the need for additional changes to that shape.
3. Solid Model Construction - The basic foundation of this design task will be the solid model of the valve, surrounding tubing and casing, and control lines. It will be built in a parametric form to hopefully allow a wide range of options for the design space. It will include some parametric design equations so that when one parameter changes, related ones will be updated.
4. Topological Optimization - The concept of topological optimization is to begin with parts that are much larger than needed and have a program remove regions that are low stressed or meet some other criteria. At some stage in the design process this approach will be utilized for some components to see what final shapes evolve. They will be compared to an existing parametric solid model for insight in revising the solid model.
5. Conduct “Water Hammer” Studies - The development of large pressure surges in fluid lines during valve closure is a well known problem. In the preliminary phases of the design evolution water hammer studies will be utilized to predict closure pressures. Later, the results from the fluid-structure interaction study are expected to yield more accurate three-dimensional pressure fields. A water hammer model will be developed utilizing Matlab.
6. Liquid Column Impact Studies - Related to water hammer studies is the knowledge that vertical piping systems can experience a second pressure surge after the valve closes. That possibility develops if cavitations cause a large gas bubble to occur above the closed valve. If so, the upward moving oil column will stop and fall back onto the valve and cause an opposite pressure load to occur. Analysis is required to determine if the valve will remain closed in such cases.
7. On Going Material Properties Surveys - Data are not currently available on the properties of the fluids likely to be passing through the SSSV in a XHPHT ultra-deepwater well application.. The oil/gas ratio is not known, so a range of likely values will be investigated. Likewise, the viscosity is not known but is important in the fluid-structure interaction calculations. Literature searches and corporate contacts will be used to develop appropriate fluid system definitions.
8. Thermal Analysis Studies - A thermal analysis of the valve assembly, control lines and surrounding reservoir will also be conducted. All of the solid and fluid components have temperature dependent properties. It has been suggested that the high temperatures can cause high annulus gas pressures to reach levels that can compromise the operation of the control systems. An investigation of predicted annulus pressures will be undertaken after the thermal studies are complete.
9. Optimization studies - Optimization studies utilizing OptDes and Ansys software will be carried out to enhance the design.
Project Start: October 14, 2008
Project End: October 13, 2010
DOE Contribution: $120,000.00
Performer Contribution: $30,000.00
RPSEA – Jim Chitwood (firstname.lastname@example.org or 713-372-2820)
NETL - Jay Jikich (Sinisha.Jikich@netl.doe.gov or 304-285-4320)
Performer Company – Ed Akin (email@example.com or 713-348-4879)
Final Project Report [PDF-7.95MB] December, 2010