"Smart" Technology Moves to Intervention Operations

Intervention jobs often get into difficulty because of a lack of accurate downhole information, despite surface indications to the contrary. For example, milling assemblies are sometimes retrieved with little to no mill wear, and overshots and spears have been pulled empty despite positive surface indications. A recent job planned for 14 runs and 30 days of rig time to recover a gravel-pack assembly from 25,000 ft, but lack of quality downhole information forced an additional 10 trips and consumed an additional 28 rig days. Without the ability to see true indications of torque, weight-on-mill, and jar-impact loads, washover shoes were pulled either untouched or junk cut, and debris was left in the well.

As wells become deeper, more tortuous, and more technically challenging for intervention, it is crucial to know more about what is actually occurring at the tool face. Traditional surface-based indicators often do not reflect the actual forces being exerted at and around downhole tools. Adding “smartness” to well-intervention operations provides valuable information on downhole parameters, including weight or tension on the tool, torque, revolutions per minute (RPM), bending stress, vibration, and pressures, that have not been seen previously at surface. Even small discrete variations can be detected. The surface information, which can be viewed from both a rig-floor monitor and a remote real-time operating center, enables better understanding of what is happening downhole. It also provides the well team with a new level of process control and real-time decision-making capability while intervention work is being carried out. Armed with valuable information and improved process control, the team can operate intervention tools with greater precision and, most importantly, with consistent reliability.

Fig. 1 illustrates the value that can be realized from smart intervention by improving performance and reducing risk exposure. The diagram shows that the cost of a well intervention is dominated by rig cost rather than service company charges and the operator’s large exposure to cost escalation during problem jobs vs. the relatively low preventive cost to mitigate disaster. Smart intervention offers a method of reducing cost and risk in that portion of a well program in which nonproductive time (NPT) is an issue.

Fig. 1—Cost comparisons between smart-well-intervention and traditional operations.

Technology Basics

Baker Oil Tools has developed a smart intervention system that integrates a modular intervention-performance sub (IPS) and a base measurement-while-drilling (MWD) tool into the bottomhole assembly (BHA). The IPS features an array of sensors, including accelerometers, strain gauges, and magnetometers (Fig. 2). All sensors are sampled at 1 kHz, and a digital signal processor inside the tool provides static parameters plus diagnostics to characterize the downhole environment. The outputs include weight on mill and mill bounce; tension at tool; torque; downhole RPM and stick/slip severity; bending stress, dogleg severity and BHA whirl; and differential pressure with equivalent circulating density. Selected parameters are transmitted uphole by means of mud-pulse telemetry. The control feedback loop complements the experienced field technician’s “sense of feel” while operating the work string at surface.

Fig. 2—Smart-Intervention BHA with base MWD tool and IPS.

Capable of receiving downlink signals, the smart system provides a link for remote control of future intervention tools. Display screens present the downhole parameters in conjunction with surface-acquired data and can be linked to a remote operations center for monitoring.

First Field Studies

Smart-intervention systems were successfully deployed in Gulf of Mexico deep water in 2005. The systems provided a surface view of downhole parameters acquired from downhole tools in milling, washing-over, fishing, casing-exit, and completion operations to confirm that actual downhole conditions in deep, high-angle wells differ significantly from what is indicated by traditional surface drilling instrumentation.

Trial 1. In the first verification application, an IPS was run in a cleanout BHA to monitor the drilling of barite, cement, and float equipment. Data transmitted to surface indicated that the milling assembly had encountered restrictions inside the casing and that circulation and reaming was required for a continued run to bottom. Strong stick/slip was encountered at 50 RPM, and rotation was increased to 80 RPM, improving drillstring dynamics and eliminating the stick/slip. Bending stress values indicated low dogleg severity in the reamed section.

Trial 2. The completion plan for an extended-reach well presented an unusual situation when a 60,000-lb load was required to hold a gravel-pack assembly in place at nearly 25,000 ft. The well geometry proved problematic because drillstring modeling showed a strong tendency for buckling. To determine the fitness of the drillstring, an IPS was used to provide downhole weight-on-bit (WOB) measurements in both compression and tension modes to verify load transfer at the end of the string. Received data indicated buckling before sufficient weight could be applied, so the drillstring design was stiffened. The IPS then showed the modified drillstring capable of transferring sufficient weight and delivered data that enabled critical decisions and facilitated a successful completion.

Trial 3. The third field trial featured the use of smart technology in a window-milling BHA. The objective was to perform a single-trip 11 7/8-in. casing exit below 11,000 ft measured depth and drill 175 ft of open hole to accept a rotary-steerable drilling assembly. By providing detailed real-time analysis of the downhole milling parameters (which differed substantially from those seen at surface), the IPS prevented an early jump-off that would have created later window-access problems. The problem was detected, and WOB was changed to eliminate that possibility. The IPS later measured the dogleg severity of the window and provided direct feedback on the operation so the rotary-steerable-drilling BHA could be run and an expandable drilling liner set through the window with confidence. NPT risk was reduced by optimizing downhole parameters, including torque, RPM, stick/slip, WOB, and rate of penetration. The downhole bending-moment data showed excellent repeatability and provided indicators for both window quality and dogleg severity. The entire operation was witnessed in real time from an onshore operating center.

Trial 4. The fourth field trial involved jarring, where an anchor latch, tubing stub, production packer, gravel-pack packer, and blank tubing were all recovered from above a gravel pack. During the anchor-latch recovery, the IPS showed a 240-lb increase in downhole weight that was not noted on the rig-floor indicator, illustrating the scale and fidelity of fish that can be detected at depth in tortuous wells. NPT was reduced by allowing the subsequent operation to be prepared before the ultralight fish was out of the hole. The job also confirmed that downhole overpull is different from surface indications and that the traditional surface-value overpull would have exceeded the limiting shear value for this operation.

Going Forward

The systems approach to smart-intervention solutions is still in its infancy, and its potential is not yet fully explored. Going forward, development efforts will focus on a bespoke surface system with a smaller footprint to improve system portability and minimize set-up efforts (Ellzey et al. 2006).

The modular architecture of the MWD system and the downlink capability open many technology-development opportunities. For example, additional sensors can be placed in intervention tools, and the tools can be actuated on command from surface by means of the downlink. Additionally, wired-pipe technology might develop into a viable communication alternative to mud-pulse telemetry, offering significantly enhanced transmission bandwidth for an even more comprehensive picture of the downhole conditions and movements with minimum transmission time delay.

In addition to future technology challenges and options, significant human-interface efforts are crucial to fully leverage the benefits this new technology offers. Intervention engineers traditionally rely on experience and “sense of feel” when selecting and operating intervention BHAs and strings. The smart-intervention-enabled process-control feedback loop that currently complements “sense of feel” will become increasingly important on the automated rigs of the future, where there is little or no sense of feel, and the field engineer must rely solely on measurements and information presented on digital displays. The feedback loop also will prove valuable in accelerating the learning curve for new operators with limited practical experience.

The smart-well-intervention system approach will drive a culture shift toward intense use of MWD technology, computers, and data management. First experience has shown that training programs, thorough pre- and post-job reviews, and guidance are required to facilitate this culture shift and encourage intervention engineers to take ownership of this technology to deliver optimal intervention performance.

Reference

Ellzey, T. et al. 2006. Innovative Systems Approach Enables Smart Intervention Solutions. Paper IADC/SPE 99121, presented at the IADC/SPE Drilling Conference, Miami, Florida, 21–23 February.

Information provided by Jim McNicol, Baker Oil Tools.