Drilling advances
The public domain is a cluttered place and simply Googling any particular subject likely yields results that are, at once, overwhelming and sorely lacking. Like oil reserves in a technically unreachable horizon, the majority of potentially valuable data floating about the worldwide web are stranded and largely inaccessible with everyday search engines.
Capturing this so-called dark data in a single-point framework is the cornerstone of a multi-dimensional toolbox the U.S. Department of Energy’s National Energy Technology Laboratory (NETL) is developing to improve the safety, environmental fitness and profitability of offshore projects. “There’s a lot more data out in the public than people realize, but it’s all over the place,” said Kelly Rose, the geology and geophysical team lead at NETL’s Research and Innovation Center in Albany, Oregon. “The majority of data online has been listed as dark data, because people don’t even know where to find it.”
More than eight years in the making, NETL’s publicly accessible Energy Data eXchange (EDX) and corresponding data visualization platforms are the springboards for science-based tools encapsulated in the evolving Offshore Risk Modeling Suite (ORM). The EDX can best be described as a virtual library of public-sourced data and resources that leverage social media appliances, but with specialized search tools for extremely complex research and energy science requirements. “All work begins with data collection, but algorithms that underpin the big data, machine learning and AI revolutions are not one-to-one transferrable into the energy and engineering industries. We’re developing new tools, like smart search algorithms, that automatically parse the entire worldwide web and brings back the things that are relevant to you,” she said on May 1, on the sidelines of the Offshore Technology Conference (OTC) in Houston. Separately, Dr. Rose noted that the DOE recently signed a two-year reciprocal agreement with Saudi Aramco to further develop the newly-patented rapid kick detection concept, which exploits previously discarded downhole measurements.
As for the ORM initiative, Dr. Rose said the enormity of elusive public data was exemplified during the recent development of a comprehensive global oil and gas infrastructure database that included wellbore, pipeline and a host of other data sets. “In less than three months, we found 700 open sources with over four million features. If you tried to find this through Google, Safari or any other search engine, you’d be challenged to find all these resources sufficiently and it would take years.”
Pre-big data. Construction of the sweeping ORM suite began with the frustrations spawned by the investigation into the Macondo oil spill in April 2010. “They (investigators) were complaining that we needed a better framework or systems to do predictions for offshore systems to understand the vulnerabilities, even from the perspective of the operator, from the vantage point of ‘did I design this well properly?,’ or ‘is this the right platform for this region?’” she said. “This was before the big data revolution. Before anyone knew what the words ‘big data’ were.”
Rose said an ORM tool under development will integrate data sets on-the-fly to enable analytical capabilities. As of now, the nine-component ORM suite includes the BLOwout and Spill Occurrence Model (BLOSOM), a 4D fate and transport oil spill simulation model, engineered to help operators understand spill dispersion tendencies. The suite also encompasses a newly developed subsurface tool that helps predict multiple subsurface properties, including pressures, temperature, fracture gradients, reservoir thickness and the like, farther afield. “What makes this different is that we use a combination of geology and geophysics, so we can predict away from where we have actual well points and data. It helps to minimize risk and improve predictions of reservoir properties, so you can do a better job of designing your well, designing your mud weight and understanding your reserve calculations.”
She said of the 150 partly ultra-deepwater wells completed since tool simulations were first run, 75% of predictions perfectly aligned with actual field data.
Quid pro quo. Meanwhile, if all goes well, within two years operators will be able to obtain exclusive licenses for the kick detection at the bit technology, clearing the way for incorporation in their drilling programs. The $750,000 collaboration with Aramco, funded through the DOE’s Technology Commercialization Funding (TCF) program, is aimed at maturing projects that are at a technology readiness level.
“When we do an agreement, it has to be mutually beneficial to both parties,” Charles Taylor, DOE global partnership partner, said at the NETL OTC booth. “There has to be something in it for the U.S. taxpayer, the U.S. government and our laboratory, and also there has to be something in it for our partners.”
Supported by a world-class in-house logging group, Aramco will bring real-world data, expertise and resources to the initiative, with the intent to integrate the finished product into its operations. Along with performing numerical modeling, Aramco will work with DOE in a field trial demonstration at one of the operator’s test sites. The DOE, in turn, will maintain patent ownership and, once the technology is deemed field worthy, can launch a full-scale promotional campaign targeting other end users. The Aramco deal addresses a lingering disconnect in moving early kick detection to the field.
“Where we’ve been stuck is we know the physics works and we know the math works,” Rose says. “The devil is in the details of how do we get this into the operational environment and how can we reconfigure these tools to make it work there. 
