Drilling advances

It takes a village to drill a well, but if the villagers have trouble speaking with each other or working together, they are not only putting themselves in harm’s way, but likely will end up paying more than they would otherwise.

That’s admittedly a simple take on an issue, that once resolved, will move the upstream community closer to capturing the full benefits of the digital age: Enabling seamless data interoperability between all the heterogeneous components, equipment and systems used in drilling a well. “One of the problems we’ve had with data is that it does not interchange,” Darryl Fett, manager of drilling and completions research for Total E&P Research & Technology USA, said of the major roadblock in sharing measurements and control data that could reduce costs, improve safety and maximize efficiency at the rigsite.

An April workshop in Houston drew more than 50 industry experts, who laid the groundwork for a possible joint industry project that aims to “create a clear and compelling path forward to greatly improve data interoperability,” which is especially critical in the drive to automate much of the well construction operation. The takeaways from the IADC-hosted workshop look to combine disparate efforts undertaken to make drilling and well data universally operable by establishing recommended practices and standards to manage the collection, transfer and processing of data.

“We don’t need a new communication protocol, but rather take the ones that are out there and standardize the way they are applied in well construction,” he told the IADC Drilling Engineering Committee (DEC) at its quarterly technology forum on June 26. “We have a lot to build on. We just need to beef it up a bit to match up to the challenges of automation.”

Scrubbing the data. Communications, however, was among the six primary issues explored at the workshop, which also looked at safety/security, quality, architecture, deployment and ownership/access. Of those, developing a structured procedure to ensure that clean, high-quality data is being transmitted from myriad permanent and temporary wellsite sensors, and aggregated into a data historian, is paramount.

“Anyone who has done machine learning knows if you start with a dirty data set, you’ll spend way more time and money—if you even do it—cleansing the data and getting it really useful. We’re doing this a lot of times with historical data, but in real time, if we can have some standard ways to aggregate or collect data, by the time it leaves the aggregator, we’ve already got a clean data set. You can’t do it otherwise in real time,” he told the forum, dedicated to “Drilling automation—where are we and where do we need to be?”

A well-documented barrier to advancing automation has been that Big Data has simply gotten too big to handle, which precludes the installation of more sensors that ostensibly could deliver useful measurements. “One of the barriers over the years is that we’re not adding more sensors, because we can’t handle the data we have now,” he said. “But, now we can with AI (artificial intelligence) and the like. We’re hungry for the data, but we don’t want to just give them data necessarily, we want to give them a dataset that will allow for timely, actionable decisions.”

Today’s drilling environment, Fett said, has accentuated the need for a standardized procedure that will increase reliance on long-string measurements and similar downhole measurement technologies. “We’ve done a lot with surface measurements in drilling. But, as some of these wells are approaching 40,000 ft (in total depth) offshore, we’re asking a lot out of tool guys and drillers to keep replying on all that surface data. Full credit to what we’ve achieved in doing that, but we need more downhole data other than angle and pressure,” he said.

While acknowledging that controlling ownership and access to data is “a big one,” Fett said it is not an issue that can be wholly tied to interoperability. The industry has long exchanged data with contracts that specify ownership and restrict access. “What makes it more challenging is that having a clean, robust dataset is more valuable and worth protecting more. But, having a way to structure all this, I think, will give you more control and allow access better,” he says.

Augmentation vs. automation. As for Total, Fett said the operator employs augmented operations, rather than full automation, across its global footprint. “It’s more versatile than automation, because when you say automation, people think robots and you push a button and drill a well. What we’re doing is augmenting it,” he said. “We’re applying a systems engineered approach that treats the well and the process of constructing and maintaining the well as a system.”

The oft-cited analogy to the automobile industry perhaps best describes what Fett means by augmenting, rather than full automation. The recent advances in adaptive cruise control technology and alerts that prevent the driver from backing into an obstacle likely hold more immediate value than the prospects of having a driverless car transport you across town.

“What they’ve really come back to is to inform the driver what’s going on. That’s what we want to give the driller or the automated process. The point we need to get to is to make sure we have the right data, the right information, on the action we take, “ he said. “The time value of that information really starts when you’re trying to prevent something that’s going to cost you a lot of money, or worse, affect the health and safety of your crew.” WO

About the Authors

Jim Redden

Contributing Editor

Jim Redden is a Houston-based consultant and a journalism graduate of Marshall University, has more than 40 years of experience as a writer, editor and corporate communicator, primarily on the upstream oil and gas industry.