The Next Big Leap

Oil and gas companies have spent years cleaning up and formatting their huge datasets for processing.

When that cleanup work is done, the industry will be able to apply machine learning, artificial intelligence and Big Data techniques to extract information and improve operations. And maybe even make exploration discoveries.

Think of it as the magic of tidying up data.

Kristina Beadle, senior consultant for Wood Mackenzie in Edinburgh, identified three key areas of advancement for oil and gas:

  • Computing power
  • Developments in artificial Intelligence and machine learning
  • Big Data

“Once you have your Big Data all ready, you can use artificial intelligence and machine learning to do some impressive analysis on it,” she said.

When Wood Mackenzie conducted its “Future of Exploration” research survey, frequently cited areas of digitalization change included advanced seismic and advanced computing power, Beadle said.

“On top of that, Big Data was mentioned an awful lot. People love data,” she observed.

Instead of being a specific data category, information scientists often define Big Data as a specialized field in computing, one that analyzes and extracts information from huge datasets. It’s an approach to working with data too large and complex for standard processing.

An essential part of Big Data analysis is getting data into a consistent and usable format, or “doing the plumbing,” said Julie Wilson, research director, global exploration for Wood Mackenzie in Houston.

“You’ve got to do this very hands-on work to digitize and clean up the data,” she said.

That’s a special challenge for the oil industry, where all sorts of data come from all sorts of places. Companies have been going through a prolonged process of getting disparate data into a compatible form for processing.

“Operators have dozens of data providers, and what they’ve been doing for years is ingesting that data” by standardizing data on one platform and using advanced computing, Wilson said.

Please log in to read the full article

Oil and gas companies have spent years cleaning up and formatting their huge datasets for processing.

When that cleanup work is done, the industry will be able to apply machine learning, artificial intelligence and Big Data techniques to extract information and improve operations. And maybe even make exploration discoveries.

Think of it as the magic of tidying up data.

Kristina Beadle, senior consultant for Wood Mackenzie in Edinburgh, identified three key areas of advancement for oil and gas:

  • Computing power
  • Developments in artificial Intelligence and machine learning
  • Big Data

“Once you have your Big Data all ready, you can use artificial intelligence and machine learning to do some impressive analysis on it,” she said.

When Wood Mackenzie conducted its “Future of Exploration” research survey, frequently cited areas of digitalization change included advanced seismic and advanced computing power, Beadle said.

“On top of that, Big Data was mentioned an awful lot. People love data,” she observed.

Instead of being a specific data category, information scientists often define Big Data as a specialized field in computing, one that analyzes and extracts information from huge datasets. It’s an approach to working with data too large and complex for standard processing.

An essential part of Big Data analysis is getting data into a consistent and usable format, or “doing the plumbing,” said Julie Wilson, research director, global exploration for Wood Mackenzie in Houston.

“You’ve got to do this very hands-on work to digitize and clean up the data,” she said.

That’s a special challenge for the oil industry, where all sorts of data come from all sorts of places. Companies have been going through a prolonged process of getting disparate data into a compatible form for processing.

“Operators have dozens of data providers, and what they’ve been doing for years is ingesting that data” by standardizing data on one platform and using advanced computing, Wilson said.

“Internally in those companies it’s already allowing much, much faster analysis of data,” she noted.

Beadle said “’acceleration’ is a key word at the moment. Everything is being done faster.”

“On the seismic, companies are able to do much more complex seismic surveys in the same amount of time they used to do something simple,” she noted. “You can screen more opportunities and identify more potential.”

But there’s more work to be done, Wilson said. The industry is still grappling with data-format obstacles while shooting for even faster processing turnarounds, especially on large seismic datasets.

“It still takes sometimes a year or two years to process all that data, and then there’s interpretation on top of that. Companies want that to be compressed to a matter of months,” Wilson said.

In pursuit of faster and more capable processing, the biggest oil and gas companies have been in something of a competition to own the most supercomputing power, according to Beadle.

“Eni currently holds the record with its 18.6 petaflops system,” she said.

In a news release last year, Eni announced it had “launched its new HPC4 supercomputer, at its Green Data Center in Ferrera Erbognone, 60 kilometers away from Milan. HPC4 quadruples the company’s computing power and makes it the world’s most powerful industrial system.

“HPC4 has a peak performance of 18.6 petaflops which, combined with the supercomputing system already in operation (HPC3), increases Eni’s computational peak capacity to 22.4 petaflops,” the company reported.

A petaflop is a thousand trillion floating-point operations per second, faster than you can run numbers even with a really good hand-held calculator.

Changing Role of Geoscientists

One result of the shift toward a more digital oil industry has been new demands on geoscientists. In general, the industry has preferred training its existing workforce in Big Data and advanced analytics rather than recruiting and retraining data scientists, Wilson observed.

“People tell me it’s much easier to train a geoscientist in data science than it is to take a data scientist and turn that person into a geologist or geophysicist,” she said.

Beadle agrees that a grounding in geoscience works out best for dealing with digital demands in oil and gas.

“As a geologist myself, I think of us as being very logical, so that kind of transition makes sense,” she said.

In addition to improving the capabilities of the industry, Wilson noted that computer-aided geoscience is providing some measure of relief for over-extended professionals.

“We’ve gone through round after round of layoffs since the downturn. A reduced workforce is being asked to do more and more,” she said.

Automatic seismic interpretation and compressive seismic imaging have been two of the emerging areas growing out of advanced computing in the oil industry.

“One of the things we hear about is automatic seismic interpretation through machine learning and artificial intelligence,” Wilson said.

“The basic picture of what’s happening in the basin can be done much more quickly, so the interpreter can zone-in on what’s really important,” she added.

Automating basic basin imaging and interpretation has allowed geoscientists to be “more creative and more effective,” Wilson said.

Compressive Seismic Imaging

Beadle cited the application of compressed sensing to seismic as an emerging technique for the industry. In medical imaging, compressed sensing has been used to enhanced sparse signals.

“ConocoPhillips has been using a new technique – they got the idea from the medical industry, from taking MRIs – called ‘compressive seismic imaging,’” or CSI, she said.

“Instead of using uniform sampling, which involves collecting data on a regular grid, CSI uses algorithms to optimize the process of selecting nonuniform data collection points. The technology enables geoscientists to reconstruct a higher-quality, more accurate picture with less data,” ConocoPhillips explained in a company publication.

In addition to signal enhancement, CSI provides environmental and safety benefits, Wilson noted.

“For ConocoPhillips, one of the other advantages of CSI is that it reduces their environmental footprint. They use it in Alaska,” she said, where it “can be quite dangerous to go out and lay these explosive charges where you need to in frontier and remote areas.”

Competition and Shared Datasets

Recent digital advances in the industry still leave plenty of room for improvement and progress. For instance, oil companies so far have been reluctant to share their datasets, Wilson observed.

“One would think that pooling data would realize the full power of unlocking data. The truth is, this is a very competitive industry,” she said, “We’re not seeing what you might call contributory datasets. People are using proprietary data.”

“In some places for service companies, like in unconventionals, where there are very large data sets in their collections, you can see how those companies are able to sort of fill in the gaps” by using well log collections and other existing information, she said.

And the industry has turned to a surprising resource for improving oil and gas exploration in the Digital Age: human brainpower.

“For small independents, when they open up a new play they very rarely do it in a place where there’s no data. There’s already data there and it’s new thinking about existing data that does it. That, to me, is very exciting,” Wilson said.

Tidying Up

Beadle said the next, big digital leap forward for the industry could come as companies finish standardizing their data and working those datasets into formats and platforms for analysis, finally cleaning up years and years of accumulated data for processing.

Yes, it’s the power of Tidying Up.

“We went from 2-D seismic to 3-D, to full-azimuth, and now we’re looking at full waveform inversion. Things have happened incrementally,” Beadle noted.

“Hopefully, there can be quite a big jump once the data is tidied up. That can open up new possibilities in machine learning and artificial intelligence,” she said.

Expect that next big jump forward to come sooner rather than later.

“The pace of change is quite fast,” Wilson said. “It’s here. It’s now. It’s happening in all these companies.”

farzwxewusybasxttwtscvyqdebqyvsxssyv

You may also be interested in ...