Riserless, Remotely Operated Technology Enables 10,000-ft Intervention Without Rig
Intervention is necessary during the life cycles of most subsea wells. However, traditional methods can make it time consuming and very costly, with spread rates for drilling and semisubmersible rigs running at USD 1 million to USD 1.4 million per day. There are more than 4,000 producing subsea (wet tree) oil and gas wells worldwide, and the number is increasing by approximately 500 per year. With many wells more than a decade old, intervention is crucial to enable maximum oil and gas extraction.
The sustained rise in deepwater exploration has made the need for cost-effective intervention on wet well trees even more pertinent. With the challenging conditions encountered in deep subsea settings such as Asia, Brazil, West Africa, and the Gulf of Mexico, many wells have produced for several years without necessary intervention. This often results in suboptimal production and reduced ultimate recovery.
The AX-S system, developed by Expro, is designed to provide a safe, riserless, and remotely operated subsea well intervention method that can complete a typical deepwater intervention in six to eight days, compared with 10 to 12 days for a rig intervention, at a cost savings of approximately 75% to 80% based on lower day rates and speedier job completion. The system, designed and built over a seven-year development period with the input of more than 200 vendors, is in the final commissioning stage. Commercialization is expected later this year.
Deployed from a monohull vessel (Fig. 1), the system is the first rigless intervention technology that can operate in waters up to 10,000 ft—several hundred feet deeper than the world’s deepest subsea well. The improved economics afforded by the system will allow operators to increase production and recovery rates in wet tree wells that otherwise might wait additional years for interventions.
To enable deployment of the system, Expro has entered into a multiyear charter party contract with TS Marine Asia Pacific to use its dynamically positioned (DP) multiservices vessel Havila Phoenix for worldwide operations. The DP 2 Class vessel is 361 ft long and 75 ft wide with a moonpool of 23.6 ft by 23.6 ft, a 94-ton fiber rope drum winch with deployment tower, a 276-ton wire rope subsea crane, and two ultraheavy-duty Work Class remotely operated vehicles (ROVs) rated to more than 13,000 ft. The winch and crane are equipped with active heave compensation systems.
Structurally, the subsea intervention system is 110 ft tall and weighs more than 242 tons. It is deployed onto a subsea tree from the vessel by means of the active-heave compensated fiber rope winch and is remotely controlled from the surface like an ROV.
The system consists of an integrated set of pressure-contained subsea packages: well control, tool storage, wireline winch, and fluid management. A hydraulic plug-pulling tool overcomes the risks of pulling and setting tree crown plugs, while a novel control umbilical overcomes the challenges of weight and subsequent deployment/handling system size. The system is effectively a fully enclosed pressure housing, with no dynamic seals between the wellbore and surrounding environments.
The well control package (WCP) is a dual safety barrier containing industry-proven 7⅜ in. shear/seal and gate valves. If safety issues arise, the operator has time to identify the problem and isolate the wellbore.
Positioned directly above the WCP is the tool storage package, which contains eight tool pockets located around its inner circumference. The tools are swapped on the seabed in minutes rather than the hours it would take on the surface. Because the tools are held in pressure-retained housing, no pressure testing is required after tool changes.
The tools are run in the well by the wireline winch package. Because of the system’s enclosed pressure housing, the risks of hydrocarbon leakage into the surrounding water and water seepage into the well are eliminated. The winch has 25,000 ft of monoconductor that conveys the various intervention tools into the well.
The final subsea section, the fluid management package, can deploy glycol fluid into the system to flush out seawater and also hydrocarbons that are circulated back into the well or, more likely, back to the host subsea production system. Depending on the specific needs of the customer, seawater can be mixed with the fluid in variable ratios, as required, for pressure testing and flushing.
A control cabin on the vessel has a computer generated interface to control and instruct the various packages on a fully automated basis. This system encompasses all hydraulic controls and these are located subsea. There are no hydraulic lines that run back to the surface. To ensure operations are safe and effective, video cameras and an ROV are an integral part of the system.
A fiber rope umbilical bundle and handling system for deployment of the technology was developed. It comprises three individual umbilicals helically wrapped around a main fiber rope and provides greater strength and operational efficiency than wire rope (Fig. 2). The bundle is buoyant in water and therefore adds no weight to the overall deployment system. This also reduces the winch power consumption and surface equipment size. There is also no torque in the lifting line, which eliminates the potential operational issue of rotating loads that often occurs with the use of conventional metallic
Back deck operations are significantly safer because the tool swaps are carried out on the seabed. To leave the location, it is only necessary to disconnect the running tool and umbilical bundle from the support vessel, which leaves the system as a full, pressure-containing safety barrier on top of the subsea tree.
The system is more cost effective than riser-based alternatives because it is supported from a vessel and is faster to operate than wire-through-water systems—especially on horizontal trees—because of the speedier tool changing process. Studies by Expro suggest that the new system may be the only economic means of wireline intervention in deepwater wells. In addition, the advantages in safety, speed, and cost saving resulting from the step change to remote subsea intervention are also applicable in shallower water.
The system is undergoing a comprehensive testing program, which started in June 2011, to ensure that every element is working effectively before it is commissioned and commercialized this year.
A three-phase wet testing process began in September. Starting at a depth of 377 ft in the Buchan Deep, east of Peterhead, Scotland, the active heave compensation, running tool, winder and umbilicals were tested. Deployment and recovery of dummy packages onto the seabed were executed, and the running tool thrusters and orientation were tested.
These elements were then successfully tested at 3,957 ft water depth in Sognefjorden, northeast of Bergen, Norway, and finally tested in deeper waters of 8,018 ft north of Shetland in the Norwegian North Sea.
The last test phase will be to install the subsea packages and run the final commissioning tests on a subsea wellhead, which is slated to be completed by the second quarter before the system is deployed on its first commercial job.
Expro has signed a long-term frame agreement with Total E&P UK, which will enable collaboration between the operator, its clients, and the service company to tailor the technology’s capabilities to anticipated operator needs. Discussions toward a similar objective are advancing with other major and independent operators.
Wet well tree intervention is increasingly important as operators look to extend reservoir life and maximize production. However, with industry budgets under scrutiny, the increasing cost and declining availability of rigs could result in a cutback of this intervention activity—assuming it is even commenced—if it is dependent on rigs.
The AX-S system, deployed from a vessel, has the potential to transform the economics of wet tree intervention and allow operators to intervene on a timely basis more closely comparable to that seen on dry well trees located on surface offshore production facilities.
Model Integration and Data Optimization Software Enables Integrated Project Modeling
Integrated asset modeling and optimization is an underserved niche in an industry where its demand is increasing because of the growing complexity of operating environments.Software called Pipe-It, which falls under the category of “integrated asset management tools,” and specifically “model integration and optimization,” was unveiled in December by Petrostreamz of Norway. It is designed for building integrated model-based tools for oil and gas asset optimization. Petrostreamz is a software spinoff of PERA, a consultancy firm in global pressure, volume, and temperature (PVT) modeling and gas enhanced oil recovery modeling.
Curtis H. Whitson, a professor at the Norwegian University of Science and Technology (NTNU) and the chief technical officer of Petrostreamz, said the software addresses a much needed, underserved segment. Although it is a commercial product, it is made available to university students and research institutes in order to cultivate a data stream translation user community in the oil and gas industry.
The company has published two Pipe-It integrated asset management (IAM) technical papers: the first in 2009 with paper SPE 121252, “Model-Based Integration and Optimization—Gas Cycling Benchmark,” and the second in 2010 with paper SPE 130768, “Multi-Field Asset Integrated Optimization Benchmark.” In both, all model data sets are made available to the public, including the Pipe-It project itself, for those who wish to test their IAM solution against the company’s published results. The second benchmark is a complex integrated model publicly documented from NTNU’s Center for Integrated Operations (www.ntnu.edu/iocenter), with specific attention to the paper SPE 130768 benchmark (www.ipt.ntnu.no/~io-opt/wiki/doku.php).
“If the application runs from a command line, you can integrate it in Pipe-It. In the past, model integration was more limited, as market solutions were sold in packages from a single vendor,” said Petrobras production engineer Sthener Campos. “It was used as a commercial strategy to tie the client. This scenario goes against a smooth implementation as it clashes with the profile of final users. Preset integration packages require users to change the software they used to develop engineering, and the knowledge they already own.”
The software performs four main functions: data stream translation, the orchestration of multiple processes, parametric analysis, and project optimization. The data stream translation is based on a technology for translating data streams for oil, gas, and water, from one description to another, with each description being needed as input for different software applications.
“The technology behind that is quite new and complicated, for example, converting from a black oil description into a very detailed compositional analysis of methane, ethane, propane, hydrocarbons, carbon dioxide, even very heavy components that need descriptions like asphaltene and other heavy hydrocarbons,” Whitson said. “It is really an area that has received very limited attention in the literature, and we have been working on the technology in this area for the last 15 to 20 years. Some is published and some is not, but the methodologies are technically transparent in the software, in the sense that they are not going to hide those methods from other partners in the field.”
Pipe-It can be mixed into the data management systems of various vendors or into homegrown applications, scripts, and various other in-house company software systems using its Linkz technology feature. That is, the system is essentially a plug-and-play data process integrator.
“People can keep working with the software applications they are working with today, but there is a need to have those applications integrated so that one application’s results are reflected in the other applications,” Whitson said. “We connect to their databases, to flat files, or to any data source going into an application, any data source that is going out of an application, or any data source being fed to another data application. Part of that is just data movement, but the technology we provide is to the engineering stream translation, needed in addition to formatting the same data in different file formats.”
When a system is processing a series of parallel data streams, the software functions to orchestrate what applications should be run first, and in what order, and what processes can run in parallel. The software performs this task, where it would otherwise require complicated manual scripting, Whitson said.
“Excel has to do this with its own logic, which it does generally very efficiently, but that is with single cells that do single equations,” Whitson said. “Pipe-It is doing the same management of performance execution, but for hundreds of applications—and the user does not have to write any code.”
The Pipe-It project comprises two XML files: the model file and the graphical XML interface file. “Once we have integrated these models, then you can do optimization of the system, either local optimization or global optimization of an entire integrated project,” Whitson said.
Different optimization problems require different solvers; that is, nonderivative-based optimization methods such as Reflection Simplex, genetic algorithms, and mapping algorithms are more appropriate than derivative-based methods for some projects, he explained.
“Our solution provides a simple application programming interface that can connect any solver to control Pipe-It. We can’t write better solvers than other people, so we’ll allow them to launch and relaunch and control Pipe-It with their solver,” Whitson said. “Other IAM vendors often use their software, but they don’t want you running other software, so most of these companies have created integration systems that work preferentially with their own software. We’re not peddling applications software, we’re peddling the means to do integrated modeling efficiently with any application they have.”
Case Matrix Module
Campos of Petrobras is finishing his doctoral research on integrated optimization at NTNU. He is using Pipe-It for his dissertation, and was involved in Petrobras’ implementation of Pipe-It at the Urucu project in the Amazon region.
In his thesis development, there are models requesting 16 inputs. Considering 2,000 production conditions, he would otherwise need to input 3,200 parameters manually, but, using a script to access the hierarchy of Hysis through the automation module, the Linkz and Case Matrix functions can be done automatically, Campos said. The results are tabulated and sent to other spreadsheet applications such as Microsoft Excel.
Pipe-It was developed over the past four years, and has already achieved success in the field with multiple field assessment optimization projects for major companies such as Petrobras and Statoil, where it has been used for complex multiple field streams integration, history matching, reserve forecasting, product blending, and product allocation and optimization.
Other reference Petrostreamz projects include Statoil’s Asgard and Oseberg areas of Norway; British Gas in Miskar, Tunisia; Cenovus in Weyburn, Saskatchewan, Canada; ConocoPhillips’ operations in Natuna Sea, Indonesia; and Saudi Aramco’s Shaybah operations.
As part of a three-year continued R&D project, Statoil has implemented Pipe-It at a commercial level in its most complex field in the North Sea, Whitson said. The Petrobras Urucu Amazon operation unit went from a three-year R&D project to full implementation last year.
The Urucu integration and optimization project was planned in 2006/2007 at the Petrobras R&D center. The R&D project was launched in June 2007 and finished in May 2009. The current status of the Urucu project has Pipe-It embedded in an in-house asset management system platform called “Gerenciamento da Produção” (GDP).
Daniel Wagner is the Petrobras process engineer who designed and coded the GDP software. The software works as a graphical interface for the Urucu optimization project. Because the interface is more scripting- and coding-oriented, the company needed something more user-friendly for the asset operational level, Wagner said. With GDP, it is able to take advantage of Pipe-It’s command-line running capabilities and XML file’s structure to automate most of the work that would otherwise have to be done manually by editing input text files
Petrobras has also developed a similar model for its natural gas compression, export, and sales system, which calculates the composition of the gas at the discharge of the compressors, based on the arrangement of the various sources connected to the suction header, Wagner said. Moving forward, gathering system routing options as part of the optimization strategy are expected at Urucu, as well as implementation of the Pipe-It in-stream compositional optimization where specific blending is necessary to optimize the liquefied petroleum gas plant output.
One of the concentrations of PERA/Petrostreamz R&D activities is to provide technology answers for development and valuation challenges of liquids-rich shale gas fields. The company has developed an extension to its Pipe-It Shale Well Optimizer to deal with condensates and volatile oils. Participation in these R&D activities makes available the existing optimization software for liquids-rich shale gas wells, and also allows user participation in the research to study the challenges posed by these resources.
The Shale Well Optimizer provides modeling capabilities in shale and ultratight gas, gas condensate, and oil resources. Three modules provide shale resource history matching, new well optimized completion design, liquids-rich pressure/volume, and PVT phase behavior fluid description. It is being developed through an R&D project that began in 2010, which continues now with Statoil. It is also being used by PERA consultants on three industrial projects in North America.
Development includes technical solutions for fluid sampling and PVT modeling; liquids yield forecasting; special well test design for liquids-rich shales; deliverability loss due to high near-well liquid saturations; relative permeability modeling; optimal economic well completion design; history matching production performance; numerical well gridding requirements; induced fracture and natural fracture modeling; and facilities design to optimize liquids recovery.
The most profitable scenarios for Pipe-It are ones that combine integration and optimization possibilities, with need of compositional fluid analysis such as blending specifications and optimization based on stream composition,
The evolution has been slow toward an open integration architecture system, which does not mean open source code, but a system that does not restrict the user on a preset of integration options, Campos said. In the past, only software from the same vendor could be integrated. The next market step is to make available integration presets between vendors. Now the trend leans toward fully open architectures.
“Nobody has thought about how to do this in an ecumenical environment,” Whitson said. “They might not use any of our software except the integrator, and all that we want is for the industry to start doing integrated modeling. We see it being used, and it can be used, and for the most part, it’s fulfilling the visions we had during the software development.”