Data aggregation and alarm systems are important to reduce drilling risk and managing resources. The ability to monitor streaming data from multiple wells and present it in a variety of forms has proliferated with the growth of real-time centers and is changing the way the industry drills wells.

However, increasingly complex wells and challenging economics, along with continued advances in technology, have many operators wondering if the mountain of data being collected cannot be put to better use.

Too often the volume of information is overwhelming. Typically its utility is confined to alarms for anticipated conditions or after-the-fact investigation. Lost in the clutter are the subtle and complex precursors to problems. It is a difficult task to see the data and pick out trends or associate multiple indicators of an impending problem.

The industry has taken note. The final report issued by the National Commission on the BP Deepwater Horizon Oilspill and Offshore Drilling states, “It is no longer acceptable to rely on a system that requires the right person to be looking at the right data at the right time and then to understand its significance in spite of simultaneous activities and other monitoring responsibilities.”

Technology is the answer, suggests the report’s authors, who note, “There is no apparent reason why more sophisticated, automated alarms and algorithms cannot be built into the display system to alert the driller and mudlogger when anomalies arise. These individuals sit for 12 hours at a time in front of these displays.”

Case-based technology

To reduce risk, improve performance, and streamline operations at its South Texas and South Louisiana drilling operations, Houston-based Swift Energy Co. is running a case-based expert system in its real-time drilling operations center (DOC) in Tilden, Texas.

Swift relies on Verdande Technology’s DrillEdge system, which uses a library of past situations or cases to identify event precursors and diagnostic indicators and assign a degree of similarity while drilling.

The information is used to predict such events as drillpipe twist-off, stuck pipe, and lost circulation. Real-time prediction is achieved through continuous comparison of the rig’s data stream with the case library. Instead of watching streaming data and reacting to alarms, personnel in the DOC are presented with information to better interpret evolving wellbore conditions and mitigate problems.

When symptoms that could lead to major events are recognized, the most relevant case histories and best practices are retrieved immediately and indicated on a circular “radar” display.

Cases enter from the outside perimeter of the screen. As the correlation with past events increases, the case moves closer to the center of the screen to convey a relative degree of importance or criticality.

Diagnostic capabilities derive from the same process. For instance, changes in standpipe pressure might indicate a drillpipe washout. The information is not predictive, but it provides information that could avoid a future twist-off.

Integration into the DOC

Before the introduction of this technology, the DOC was collecting data using various rig sensors. The data were fed to a WITSML server and then to a data aggregation service in preparation for display. The process functioned as it was designed and worked fine for anticipated problems and direct issues. But it was not providing the answers to situations that were more complex and harder to diagnose, such as those requiring input from multiple parameters and variables.

The software was introduced into the DOC for an initial 30-day trial and was manned by Verdande Technology personnel. During this trial, drills were run to better understand system response and identify benefits the technology could provide to Swift personnel to improve decision-making.

The primary objective of the trial was to demonstrate the utility and functionality of the DrillEdge system. It was important that Swift personnel have full confidence in the system’s ability to deliver relevant information when challenging drilling situations arose related to nonproductive time (NPT)- causing events.

After a successful initial trial, a second phase of testing is under way. Current operations are running the system without vendor oversight, which has reduced the number of people in the DOC and increased the confidence Swift personnel have in the technology. An ongoing training process prepares engineers to interpret cases and develop new ones as situations warrant, with the goal of improving overall system response.

Further efficiency is anticipated with rig personnel receiving alerts based on cases appearing on the system’s radar screen. Enabling the drillers frees DOC engineers to look at more wells using fewer screens, which allows them to focus efforts on the situation or well that requires dedicated attention.

DrillEdge software offers a radar interface that allows personnel to easily collaborate on multiple wells. (Images courtesy of Verdande Technology)

Escalation process

A key aspect to the performance of the expert system is response. Once a case is on the radar, there must be a process in place for resolving the situation.

An escalation and notification protocol has been developed so that Swift personnel manning the DOC can provide the right information to the rig when problematic situations are identified and remedial actions (based on experience and company best practices) are presented by the technology.

The process determines a series of steps that first confirm the data and then monitor or escalate the situation to a higher degree of concern. When the case similarity match is between 50% and 75%, it is monitored. If the situation is static or worsens over a 10-minute period, it is escalated.

A case similarity match exceeding 75% is immediately escalated to a hierarchy ranging from the rig personnel to the DOC superintendent and on-call engineer.

If changes are made that improve the situation, the information is entered into the log, and monitoring resumes until the case exits the radar. Otherwise the cycle continues, and other remedies are applied until the situation is resolved. DrillEdge captures all of this information for reference the next time similar conditions occur.

Diagnostic support

A similar process of matching events to problems is being explored to provide diagnostic support. In this application, observed events lead to the identification of a current problem.

For instance, fluctuations in standpipe pressure exhibit an identifiable fingerprint that can indicate a washout in the drillstring. The appearance of such a case on the radar notifies the driller so action can be taken immediately to prevent a possible twist-off.

The system has been trained to recognize conditions relevant to drillstring washouts, and it will immediately present a succession of events. In real time, the pressure loss and a washout event on the radar screen would ultimately cross an escalation threshold, and a response would be initiated in time to prevent a twist-off from occurring.

In addition to drillpipe washouts, this diagnostic capability is being examined in regard to several other applications involving downhole tools, including drill bits and mud motor stalls that lead to failure.

Advancing the process

To avoid drilling problems before they arise, operators are making greater use of rig data streams. Building on current infrastructure and industry services, Verdande’s unique system provides advanced detection and decision-making support. Case-based technology extends data utility far beyond traditional monitors and alarms to enable real-time problem prediction and diagnostics for reduced risk and greater efficiency.

Cases on the radar are linked to customer-defined recommended mitigation practices.